 Your new post is loading...
|
Scooped by
Charles Tiayon
September 9, 1:50 PM
|
Avec la traduction, le dispositif franchit une étape autrement sensible. GoRead décrit une succession d’agents : première version, traitement séparé des noms, lieux et expressions locales, reprise des formulations signalées comme incertaines, révision linguistique, puis arbitrage contextuel. "Trois mille couronnes suédoises — environ 270 € — pour faire passer un roman de 80.000 mots dans une autre langue. GoRead, la plateforme d’autoédition cofondée par Jonas Tellander, cofondateur de Storytel, ajoute la traduction automatique à sa chaîne de production. Neuf langues sont proposées, sans relecture humaine systématique. Un tarif qui change l’échelle économique de l’opération, sans faire disparaître les questions de droits. À ce prix-là, GoRead change sérieusement l’échelle annoncée. Selon Boktugg dans son enquête du 7 septembre 2026, le service permet désormais de traduire un manuscrit entre neuf langues : suédois, danois, finnois, islandais, anglais, allemand, polonais, néerlandais et bulgare. Pour un roman d’environ 80.000 mots, la facture tourne autour de 3000 SEK (Couronne Suédoise) par langue, soit quelque 270 €. Le service de publication de GoRead, lancé en février 2026, aurait déjà fait paraître 250 titres, principalement en Suède, indique Jonas Tellander. La société avait jusque-là concentré son automatisation sur la fabrication du livre : contrôle orthographique, proposition de quatrième de couverture, couverture, EPUB, fichier pour l’impression et narration audio. Sur son propre site, elle revendique 70 % des revenus reversés à l’auteur, qui conserve ses droits sur l’ouvrage. Avec la traduction, le dispositif franchit une étape autrement sensible. GoRead décrit une succession d’agents : première version, traitement séparé des noms, lieux et expressions locales, reprise des formulations signalées comme incertaines, révision linguistique, puis arbitrage contextuel. « Cela fonctionne vraiment bien lorsque les différents agents examinent le travail du précédent », assure Tellander. Mais aucune vérification manuelle n’est effectuée par GoRead. L’auteur peut faire relire le résultat avant publication — à ses frais et à son initiative. Le tarif porte donc sur la génération et le traitement automatisé du texte, pas sur une validation humaine. Neuf langues, et l’audio avec
L’extension linguistique vaut aussi pour le livre audio. GoRead annonce pouvoir produire des narrations synthétiques dans les mêmes neuf langues et distribuer ebooks et audiolivres sur les marchés couverts par Storytel, BookBeat, Nextory et Audible. Au lancement du service, les voix artificielles étaient produites avec la technologie d’ElevenLabs. C’est du côté d’Audible que l’information mérite une précaution. Boktugg affirme que GoRead est parvenu à faire accepter ses narrations produites par IA, alors que le service n’acceptait jusque-là pas, selon le média suédois, les voix artificielles provenant de producteurs extérieurs. Aucune annonce publique d’Audible consultée ne permet cependant d’en conclure à une ouverture générale de sa plateforme. Le développement doit donc rester attribué à GoRead et à Boktugg. Spotify devrait suivre, probablement au cours de l’automne. Pour le papier, l’internationalisation demeure plus prosaïque : GoRead continue de travailler avec Publit et ne distribue actuellement les exemplaires imprimés hors de Suède qu’en Finlande et au Danemark, les frais de transport limitant une extension plus large. Un bouton « traduire » ne cède pas les droits
Reste la question que 3000 couronnes ne règlent pas : a-t-on le droit de faire traduire le texte que l’on vient de déposer ? En Suède, rien n’interdit en soi de recourir à une IA pour traduire un ouvrage. En revanche, l’article 2 de la loi suédoise sur le droit d’auteur réserve au titulaire des droits l’exploitation de l’œuvre, y compris « en traduction ou adaptation ». Un écrivain qui possède encore ses droits de traduction peut donc recourir à un tel outil ; celui qui les a cédés doit d’abord regarder son contrat. La promesse commerciale de GoRead — l’auteur conserve ses droits face au service — ne répond pas à cette question préalable. Elle décrit la relation avec GoRead, pas les engagements éventuellement conclus avec un éditeur, un agent ou un autre partenaire. Un second problème apparaît une fois le texte traduit. L’article 4 de la même loi reconnaît des droits à celui qui traduit ou adapte une œuvre, sans lui permettre de porter atteinte à ceux attachés à l’original. Mais le PRV, l’office suédois de la propriété intellectuelle, rappelle que la protection suppose un créateur humain capable d’effectuer des choix libres et créatifs. La Cour de justice de l’Union européenne applique elle aussi ce critère d’originalité. Pour une version générée intégralement par une machine, sans intervention créative humaine suffisante, l’existence d’une protection autonome demeure donc incertaine. À l’inverse, une reprise substantielle par une personne peut introduire des choix susceptibles d’être protégés. Aucun texte ni aucune décision identifiés ne tranchent spécifiquement le cas des traductions produites par GoRead. L’AI Act européen ne rend pas davantage le service illégal. Son article 50 prévoit notamment, pour les fournisseurs de systèmes générant du texte ou de l’audio synthétique, des obligations techniques de marquage des sorties, assorties d’exceptions. Mais l’obligation faite aux utilisateurs de signaler un texte produit ou manipulé par IA vise les contenus publiés afin d’informer le public sur des questions d’intérêt public : elle ne crée pas, à elle seule, une mention obligatoire « traduit par IA » sur chaque roman. GoRead ne précise pas publiquement quel modèle assure ses traductions. Impossible, dans ces conditions, d’attribuer plus précisément les obligations entre les différents acteurs techniques. Pour les auteurs autoédités — et, demain, les petites maisons auxquelles GoRead veut également ouvrir ses outils — la barrière économique baisse brutalement. La responsabilité éditoriale et juridique, elle, reste entre des mains très humaines.
Nicolas Gary
09/09/2026
Contact : ng@actualitte.com
https://actualitte.com/article/133775/economie/pour-3000-couronnes-goread-traduit-un-roman-par-ia
#metaglossia
#metaglossia_mundus
"Charles Tiayon, cartographer of the translation world
Charles Tiayon is a translator, lecturer and consultant whose Scoop.it activity reflects the professional universe named in his bio: translation, terminology, lexicography and intercultural issues. Through METAGLOSSIA, his public selection tracks language as a field of work, a field of rights, and a field being altered by AI.
A systems view of language
Tiayon looks at translation as a living system. His selections connect interpreters, translators, dictionary makers, universities, publishers, language activists and software companies, rather than treating them as separate worlds.
Technology with memory
AI is a recurring concern, but it is not treated as the whole story. Real-time speech translation, OCR and voice AI appear alongside older and slower forms of linguistic work: dictionaries, terminology, literary translation and language education.
A multilingual compass
The geographic spread is notably international, with attention to French, Spanish, Chinese, Arabic, Maltese, Breton, Ukrainian, Filipino, Indigenous languages in Canada and Indian language politics. No single country dominates the map.
What defines this selection
Archive-scale attention
With 99,235 posts analysed in a single topic, this is a long-running, high-volume observatory of the translation world. The account dates back to 2011, giving the selection the feel of an archive as much as a live feed.
A broad professional lens
The material moves easily between AI translation tools, national dictionaries, literary translation, terminology work, market news and language policy. That range reflects a professional view of language work as both craft and infrastructure.
Mixed-source reliability
The sources span institutions, universities, technology companies, mainstream media and regional outlets. UNESCO, university sites, DeepL, Mistral, CBC, Ouest-France and other sources create a mixed diet of policy, research, product news and local reporting.
Specialist reach, public scale
The audience numbers are substantial, with more than 567,000 views and over 424,000 unique visitors across all topics. The reach suggests sustained interest beyond a narrow specialist circle.
Topics
◍
Metaglossia: The Translation World
This flagship topic covers translation, interpreting, intercultural communication, terminology and lexicography as they unfold in public life. It is especially strong on AI language tools, institutional language policy, minority languages, literary translation and dictionaries.
99.2K posts
en
Deep dives
Where language becomes infrastructure
Charles Tiayon’s main topic reads like a newsroom for the language professions, but with a wider lens than trade news alone. Translation, interpreting, terminology and lexicography appear beside minority-language rights, literary circulation, dictionaries, competitions, and the fast-moving world of AI speech and OCR tools.
Metaglossia: The Translation World
The strongest thread is the collision between language as culture and language as system. A new Maltese digital dictionary, a Doha historical dictionary with text-to-speech, UNESCO and Unicode cooperation, and First Nations language translation all sit in the same frame as DeepL, Mistral OCR, and real-time speech models.
Metaglossia: The Translation World
His editorial voice is not silent aggregation. The analysis points to regular INSIGHT notes, often substantial and quote-based, with an explanatory tone and occasional sharper framing when language politics or AI consequences are at stake. The result is a practical map of a field being reshaped by institutions, software, markets and communities at once."
Courtesy
Scoop.it
06.07.2026
#metaglossia_mundus
#metaglossia
Alphabétisation : 23 millions de Congolais ne savent ni lire ni écrire, selon le Gouvernement | Radio Okapi
"Alphabétisation : 23 millions de Congolais ne savent ni lire ni écrire, selon le Gouvernement
Publié le mer, 09/09/2026 - 12:16 | Modifié le mer, 09/09/2026 - 14:14
Kinshasa analphabétisme Eva Bazaiba Société Actualité National
La République démocratique du Congo compte au moins 23 millions de personnes ne sachant ni lire ni écrire, soit un taux d’analphabétisme de près de 20 %, selon les chiffres présentés mardi 8 septembre 2025 par la ministre d’État en charge des Affaires sociales, Ève Bazaiba.
Elle s’exprimait à l’occasion de la 60e édition de la Journée mondiale de l’alphabétisation, célébrée autour du thème : « Alphabétisation pour le peuple, la planète et la prospérité ».
Les conflits et les crises affectent l’accès à l’éducation
Selon le gouvernement, cette situation est notamment liée à la persistance des conflits armés, aux crises épidémiques et aux chocs climatiques, qui provoquent des déplacements massifs de populations.
Dans sa déclaration, Ève Bazaiba a souligné les conséquences de ces déplacements sur l’accès à l’éducation.
« Arrachées de leur cadre de vie habituel, les personnes déplacées internes et les réfugiés subissent des conditions d’existence particulièrement précaires qui compromettent durablement leur accès à l’éducation. »
Les personnes déplacées et les réfugiés font ainsi partie des populations particulièrement exposées aux difficultés d’accès à la scolarisation et aux programmes d’alphabétisation.
Des programmes de rattrapage et de formation
Pour répondre à cette situation, le gouvernement indique avoir engagé plusieurs actions avec les acteurs du secteur. Elles portent notamment sur l’actualisation et le renforcement des programmes de rattrapage scolaire et d’alphabétisation.
L’exécutif prévoit également la production de supports pédagogiques et andragogiques ainsi que la professionnalisation des éducateurs sociaux.
Le développement des compétences professionnelles et la recherche de solutions adaptées aux besoins des populations vulnérables figurent également parmi les mesures annoncées.
La Journée mondiale de l’alphabétisation s’inscrit dans le cadre de l’Agenda 2030 des Nations unies et de l’Objectif de développement durable (ODD) n°4, consacré à une éducation inclusive et équitable." https://www.radiookapi.net/2026/09/09/actualite/societe/alphabetisation-23-millions-de-congolais-ne-savent-ni-lire-ni-ecrire
#metaglossia
#metaglossia_mundus
New Translation of the Epistles Named Australian Christian Book of the Year
"A new translation making the New Testament epistles accessible to children has been named the 2026 Australian Christian Book of the Year, with judges calling the project "unprecedented".
The award was announced at the SparkLit Awards Night on Thursday 20 August, alongside the Australian Christian Teen Writer Award, won by Hanna Pheh for The Unintended Race.
The Australian Christian Book of the Year Award has recognised excellence in Christian writing since 1981 and carries a 3000 dollar prize for the winning author.
A children’s translation of the New Testament epistles has taken out the country’s top honour for Christian writing, with judges praising its ability to bring ancient Scripture to life for readers of all ages.
God’s Very Special Letters: The Epistles for Kids – covering Philippians, Colossians and Philemon, and 1 Peter – has been named the 2026 Australian Christian Book of the Year. The announcement was made at the SparkLit Awards Night, held on Thursday 20 August at St Alfred’s Anglican Church in Blackburn North, Melbourne.
A fresh way into familiar Scripture
The book is a collaboration between Matt Dodd, a pastor at Lakemba Anglican Church, and illustrator and cartoonist Thom Bransdon, published through The Open Word.
Judges praised the translation for its direct, plain language, noting it’s “a joy to read” and that the accompanying illustrations advance the narrative, reflect the culture of the original readers, and keep people at the centre of complex theological ideas.
Judges called the project “unprecedented”, posing the question many readers may find themselves asking:
“This unprecedented and wonderful initiative makes these epistles accessible to children and offers older readers a fresh encounter with familiar and favourite parts of Scripture. Why hasn’t this been done before?”
Why the epistles, and why now
Speaking to Hope 103.2’s Ben McEachen last year, Matt explained the thinking behind the project and why it doesn’t take the easy road of simply paraphrasing an existing children’s Bible.
“There have been many attempts to translate the Bible into the language of children,” Matt said.
Hope 103.2 is proudly supported by
“There also have been many attempts to adapt the message of the Bible into kids’ resources that look and feel like kids’ books. But there really hasn’t been an attempt to do both at the same time.”
Matt and Thom chose to begin with Philippians, Colossians and 1 Peter partly because the epistles can be overlooked as biblical texts for younger readers.
“The letters in the New Testament are windows on to the lives of real Christian people being transformed by the gospel, diverse communities of young and old coming together around Christ,” he said.
“We want to make books for children that are full of colour and adventure, because the Epistles are full of colour and adventure.”
For Matt, the goal goes beyond readability. He wants children to be able to encounter God’s word for themselves, without needing an adult to translate or explain it along the way.
“[They] can have a personal relationship with God without needing the help of adults,” he said.
“They can relate to God, from the word of God themselves, without barriers, stress or pressure. I can hear the voice of my heavenly Father; he can be my father; I can be his child. That’s the passion; that’s what is driving us.”
A night of celebration for Christian writers
The Australian Christian Book of the Year Award has been presented annually since 1981, recognising original writing by Australian citizens that contributes to Christian literature in Australia. The winning author receives a 3000 dollar prize, along with a framed certificate for both author and publisher.
This year’s shortlist of nine included a wide range of voices and genres, from Stan Grant’s reflections on faith and identity in Murriyang, to Jessica Claire Bond’s A King Washes Feet, an exploration of grace inspired by her popular footwashing series.
The same night also saw Hanna Pheh, from Mount Waverley in Victoria, awarded the 2026 Australian Christian Teen Writer Award for her novel The Unintended Race, which follows a young girl wrestling with questions of life, death, forgiveness and divine providence as her family comes to faith, set against the backdrop of 1980s Kuala Lumpur.
Encouraging the next generation of Christian writing
This year’s judging panel – Greg Clarke, Amy Isham and Grace Lung – read and assessed entries on their originality, literary style, design, and contribution to Christian writing in Australia.
Whether you’re after a fresh way to introduce your children to Scripture, or simply want to see familiar passages in a new light, God’s Very Special Letters is available now through The Open Word.
Congratulations to Matt Dodd, Thom Bransdon, Hanna Pheh, and everyone shortlisted this year."
By Hope 103.2
8 September, 2026
https://hope1032.com.au/books/new-translation-of-the-epistles-named-australian-christian-book-of-the-year/
#metaglossia #metaglossia_mundus
The Axon body camera can translate more than 50 languages during calls
The sheriff's office says the technology is over 90 percent accurate in translating
"LEE COUNTY, Fla. — The Lee County Sheriff's Office is equipping deputies with real-time translation technology for body cameras.
The new artificial intelligence feature is designed to improve communication in the field. The Axon body camera can translate more than 50 languages during calls.
Advertisement
LCSO started rolling out the new Axon body cameras in May. Sheriff Carmine Marceno says a series of federal grants paid for them this year.
"This is going to be the body cam of choice," said Marceno.
When translation is activated, the body camera records both audio and video. The sheriff's office says the technology is over 90 percent accurate in translating.
Sheriff Marceno says it's been a game-changer for deputies in the field.
"If I pull up to a residence, someone is a Spanish-speaking person, I can talk to them and speak to them in Spanish and understand exactly what their needs are, and they can understand what mine are," said Marceno.
For those who struggle with getting information across in English, Margot Aguilar says this new feature will make interactions with deputies more comfortable.
"Well, I think it's really important for all the community of Latinos. The translation is good for interacting with police, so that there's no confusion for those who don't know the language," said Aguilar (translated to English)."
Updated: 5:35 PM EDT Sep 8, 2026
Editorial Standards
Madison Zaleski
Reporter
https://www.gulfcoastnewsnow.com/article/lee-county-deputies-get-real-time-translation-body-cameras/73644531
#metaglossia
#metaglossia_mundus
"Google has introduced major upgrades to Live Translate in its Google Translate app, bringing real-time, two-way conversational audio and on-screen text translation to Android and iOS users.
According to Google, the upgrades are designed to make conversations across languages more seamless, with Live Translate now supporting more than 70 languages, including Hindi, Arabic, French, Korean, Spanish and Tamil.
A key addition is Gemini-powered speech intelligence. Google says its advanced speech recognition and multimodal Gemini models can help isolate background noise in environments such as busy cafés and airports, while also recognizing accents, intonation and conversational pauses.
Live Translate also provides dual-language transcripts on screen while playing the translated audio, allowing users to follow conversations both visually and through sound. The app can automatically detect the active language as users speak.
Google has also added a companion language-learning feature that creates customized speaking and listening practice sessions based on individual learning goals.
How to use Live Translate:
1. Open the Google Translate app on an Android or iOS device.
2. Tap the Live translate option.
3. Select the source and target languages.
4. Start speaking naturally. The app automatically detects the active language and displays and reads out the translation.
Google has also announced two additional Live Translate upgrades. On Android, Live Translate can process conversations in the background, including when the phone's screen is locked, allowing users to multitask. Support for this functionality on iOS is coming soon.
On iOS, translated audio can now be played through the phone's earpiece. This complements the existing Android support for hands-free listening, particularly in noisy environments such as airports and public transport.
The latest changes are aimed at making real-time translation more practical for travel and everyday conversations, while expanding the ways users can listen to and follow translated speech within the Translate app."
Story by ANI | Posted by Ashhar Alam | Date 08-09-2026
https://www.awazthevoice.in/gadgets-news/google-translate-gets-real-time-two-way-translation-upgrade-67175.html
#metaglossia
#metaglossia_mundus
Accelerating its production of English-language translations of popular genre titles from France, Germany, Italy, and Spain to meet growing demand
European translations will account for nearly 25% of its slate of English-language Originals, which includes audiobook content that Audible commissions, produces, or releases exclusively for its own platform
"Audible says it has "quadrupled" the number of titles it has translated into English from European languages so far this year, with translations from the continent up 300% year-over-year. (The company did not provide a comparative figure from last year).
According to the company, by the end of 2026, European translations will account for nearly 25% of its slate of English-language Originals, which includes audiobook content that Audible commissions, produces, or releases exclusively for its own platform. The titles primarily come from Germany, France, Spain, and Italy. A spokesperson for Audible confirmed that all translations were done by humans, not AI.
All this is being done, the company said in an announcement, to meet demand for genre titles, including crime, thriller, mystery, and romantasy. German-language titles are among the most popular, including Anthony Khaseria's The Ratcatcher 2, and Elli J. Morrigan's romantasy sequel Forged By Fire, which follows Bound By Fire, Audible's most-listened-to English translation to date.
"The success of these titles has reinforced our belief that great storytelling transcends language and borders," Aurelie de Troyer, Audible's head of content, international, said in a statement. "As audio becomes increasingly global, we're able to introduce English-speaking listeners to exceptional writers and creators through a new medium.""
By Ed Nawotka |
Sep 08, 2026
https://www.publishersweekly.com/pw/by-topic/international/international-book-news/article/101185-audible-ramps-up-english-translations-of-european-audiobooks.html
#metaglossia
#metaglossia_mundus
UN resolves on Africa's true size on the world map
164 member states vote in favour, the US votes against, while six other countries — Serbia, Estonia, Georgia, Lithuania, Moldova and Ukraine — abstain
"The United Nations General Assembly endorsed a new world map that represents Africa more accurately in its larger size, a culmination of a recent campaign led by African countries and advocacy groups.
Friday’s resolution, which was sponsored by Togo and backed by members of the African Union, was passed with 164 member states voting in favour. Only the US voted against it while six other countries — Serbia, Estonia, Georgia, Lithuania, Moldova and Ukraine — abstained.
The vote encourages governments and institutions around the world to replace the traditional Mercator map, commonly used since the 16th Century, with the Equal Earth projection that advocates and geographers say shows locations more accurately.
The Mercator projection shows Greenland and Africa as appearing to be about the same size whereas in the Equal Earth projection that shows continents in their true proportions, Africa is 14 times larger than Greenland.
Friday’s resolution does not ban the Mercator projection or impose a replacement, but only encourages the use of the Equal Earth projection, the UN said in its report.
“Maps shape our understanding of the world,” the global body quoted Togo’s Foreign Minister Robert Dussey as saying before the vote. “They guide education, nourish the imagination, and influence collective perceptions.”
Africa No Filter, one of the lead campaigners for the adoption of the new map, hailed the resolution, adding: “We want to see accurate maps in schools, textbooks, newsrooms, businesses and on the digital platforms billions of people use every day.”
Several countries hailed the adoption and urged the adoption of the Equal Earth map. “Changing the map obviously doesn’t change the world. But correcting a representation that distorts it is already an act of truth,” French Foreign Minister Jean-Noël Barrot said on X after the vote.
The Mercator map was designed by Flemish cartographer Gerardus Mercator to help European navigators at sea, but advocates say it distorted landmasses by enlarging regions near the poles such as North America and Greenland while shrinking Africa and South America.
The 2018 Equal Earth projection, on the other hand, follows the Earth’s curvature and shows continents in their true proportions.
Geographers say the traditional map is only useful for navigation.
“Outside of that very narrow navigation application, there is no point in using it,” according to Mark Monmonier, a Syracuse University professor of geography.
New world map reflecting Africa's true size endorsed by UN https://share.google/pCJOBZCCFBkm3PDJP
#metaglossia
#metaglossia_mundus
Not just language conversion but cultural transmission”: comparing human-first and AI-first workflows for translation quality, efficiency, and speed in the Anthropocene
"Artificial intelligence (AI) has reshaped translation workflows, moving from manual, sentence-level crafting to integrated pipelines using translation memories, neural machine translation, and large language models, yet education struggles to prepare translators for these shifts, risking deskilling and ethical challenges. This study aimed to optimize training by comparing two AI-assisted workflows in a food-related menu translation task, evaluating quality, time, edits, and trainee perceptions to enhance efficiency, accuracy, and self-efficacy. Class 1 (n = 29) used an AI-first workflow, where AI generated initial drafts followed by human post-editing to correct errors and refine outputs. Class 2 (n = 27) employed a human-first workflow, with translators drafting texts and AI polishing to enhance style and coherence. Quality (fluency, adequacy), time, edit behaviors (total, necessary, harmful), and perceptions were assessed via quantitative metrics and qualitative feedback. Class 2 outperformed Class 1 in adequacy (M = 43.02 vs. 39.22, p = 0.002, d = −0.85) and total score (M = 89.93 vs. 86.67, p = 0.012, d = −0.68), with no fluency difference. Class 1 was faster (M = 2763.69 s vs. 3494.11 s, p < 0.001, d = −2.40) but made more harmful edits (d = 1.07). Class 1’s time-quality correlation (r = 0.905, p < 0.001) showed effort-intensive post-editing, unlike Class 2’s efficiency. Human-first workflows improve quality and confidence, while AI-first prioritize speed. Training should emphasize human-first workflows, prompt design, and selective post-editing to align with workplace curation roles, countering deskilling."
Kefang Chen
Humanities and Social Sciences Communications (2026)
Abstract
https://www.nature.com/articles/s41599-026-08906-1
#metaglossia
#metaglossia_mundus
"48% of Localization Isn't Translation, and AI Can't Do That Part
2002 Studios Media CEO, Steven Gurevitz, explains why faster translation doesn't always lead to better localization.
Translation services are expected to account for more than 52% of the global localization strategies market by service type this year, according to Coherent Market Insights, making them the industry's largest service category.
That leaves 48% of the spend going to everything else.
Demand is growing across the wider language services industry, too. Fortune Business Insights projects the global language services market will increase from $81.5 billion this year to $147.5 billion by 2034, a compound annual growth rate of nearly 8%.
But Steven Gurevitz, CEO of 2002 Studios Media, a leading content production and localization agency, says the larger share of that spend buys the least valuable part of the job.
"A lot of the spend on localization, which is translation, is what I call churn spend. In fact, no one's making money out of it. It's being fed into AI credits, and nowadays it's a commoditized item. The bit where the value is in the 48%," Gurevitz says.
AI can now translate huge volumes of content in seconds.
In fact, 88% of content decision-makers say their organizations already use generative AI for translation in some form, and 70% of translations are now machine-assisted, according to Forrester's Rethinking Localization in the Age of AI.
But customers experience much more than text. They move through websites, apps, videos, and support material, all of which need to work naturally in their language and culture.
The team at 2002 Studios Media believes that's where many localization projects begin to fall apart, because businesses often underestimate the engineering, production, and testing that follow translation.
Gurevitz adds that even with the translation speed offered by AI, the amount of work needed to localize digital products hasn't gone down.
"Customers aren't reading isolated blocks of text. They're using websites, apps, videos, and software. Every part of that experience has to work naturally in their language."
Why Businesses Keep Putting the Translation Process First
The misconception that localization means translation isn't new.
Much of the localization industry grew around translation services, so linguistics naturally became the starting point for many projects.
"The industry was set up by translation companies. They're often who people are going to. They've got the most money, they do the customer acquisition. But you're essentially getting into a food chain where we're at the back end," Gurevitz says.
Unfortunately, this resulted in production planning, engineering, multimedia, and quality assurance being treated as separate stages that come later.
Problems often begin before translation starts. Websites, videos, and e-learning courses are sometimes sent for translation before the source material has been properly prepared.
As layouts or technical requirements change, translators, designers, developers, and editors can find themselves revisiting work that could have been done correctly the first time around.
This is also seen when it comes to video localization.
A 30-second television commercial usually needs every translated voiceover to match the original runtime. A two-minute training or promotional video often doesn't because viewers won't compare different language versions side by side.
Yet many organizations still request tightly synchronized audio for every project because those selling and managing these solutions often don't understand the variables and options available to them, given that their core expertise is linguistics.
This often adds production time without necessarily improving the final experience.
"Many localization problems are created long before anyone starts translating," Gurevitz says.
"If nobody questions the project scope or workflow at the beginning, every team that follows spends time fixing problems that didn't need to exist.""
Ryan de Smidt
Sep 03 2026
https://news.designrush.com/ai-translation-localization-2002-studios
#metaglossia
#metaglossia_mundus
"For Egyptian translator and scholar Yahya Mokhtar, Chinese began as a university subject. It later became a seven-year journey in China, leading to more than 30 translations, major awards and a career connecting Egyptian and Chinese cultures.
Egyptian scholar and translator Dr. Yahya Mokhtar, who teaches Chinese literature and translation at Suez University, has translated more than 30 Chinese literary, academic and cultural works into Arabic. (Photo/Ayman El-Kady, Bridging News Cairo Bureau)
Cairo - When Yahya Mokhtar first enrolled in the Chinese language department at university, he was beginning what seemed like a conventional academic journey: four years of studying a language that was geographically distant, linguistically unfamiliar and culturally very different from his own.
He did not know then that Chinese would eventually take him far beyond the classroom.
It would take him to Beijing, where he would spend seven years pursuing master’s and doctoral degrees, introduce him to a rapidly transforming China, and eventually turn him into an academic and translator whose work has carried Chinese literature and culture into the Arabic-speaking world.
Today, Mokhtar teaches Chinese literature and translation at Suez University and works as a translator of Chinese literary, academic and cultural works.
Over the years, he has translated more than 30 works from Chinese into Arabic and received recognition in both Egypt and China for his work.
But for Mokhtar, the journey began with a decision to go further than simply learning the language.
From studying Chinese to living in China
“I started my relationship with the Chinese language during university, when I joined the Chinese language department,” Mokhtar told Bridging News.
He studied Chinese in Egypt for four years before deciding that university classes alone would not be enough.
He began studying independently, preparing for the HSK, the standardized Chinese language proficiency examination.
After eventually achieving the highest level, he earned a scholarship to pursue a master’s degree at Beijing Language and Culture University, one of the world’s best-known institutions for the study of Chinese language and culture.
The master’s degree opened another door. While studying in Beijing, Mokhtar succeeded in securing another scholarship for a doctoral degree, allowing him to complete both degrees in China.
In total, he spent seven years there. Those years were more than an academic experience. From around 2010 to 2017 or 2018, Mokhtar witnessed China undergo rapid economic, technological and social change.
“I witnessed with my own eyes the enormous economic development taking place in China,” he said.
He also saw what he described as the commitment and attention to detail in everyday life. For the young Egyptian student, China became more than a subject to study; it became a society he could observe from within.
Mokhtar’s memories of China extend beyond universities and classrooms.
As a foreign student, he received extensive support with administrative procedures, academic matters and personal challenges.
“We found all kinds of support and assistance from the Chinese people,” he said, adding that the experience left a lasting impression on him.
That experience continued to shape how he viewed China after returning to Egypt and influenced his approach to his work.
Mokhtar did not study Chinese culture from a distance. He lived in China, learned the language, studied its literature and witnessed the country’s rapid transformation—experiences that later became central to his work as a translator.
Bringing Chinese literature Into Arabic
Today, Mokhtar is a lecturer in Chinese literature and translation at Suez University, while also working as a translator of literary, academic and cultural works.
His translations have covered novels, poetry, studies and reference works, including Beijing Beijing, The Black Horse, River of Time, 1942, Meeting in Beijing, China at the Forefront, The Chinese Islamic Encyclopedia and works examining the cultural and economic dimensions of the Belt and Road Initiative.
He has also participated in cultural exchanges between Egyptian and Chinese writers and intellectuals, as well as conferences and workshops involving literary figures from both countries.
His work has extended beyond written translation. He has participated in interpretation and joint cultural activities and previously taught Arabic at Beijing Language and Culture University.
Among the Chinese literary figures with whom he has crossed paths is Nobel laureate Mo Yan. His achievements have brought recognition from both sides.
He has also received several honors in China, including the Wang Zengqi Literary Writing Award and recognition from the China Scholarship Council.
In Egypt, he won the Best Translated Book Award at the 2022 Cairo International Book Fair for his translation of the encyclopedia Economies of the Belt and Road Countries and Regions. The award was presented by Egypt’s Ministry of Culture and National Center for Translation.
He has also been recognized in China for his contributions to books and literary exchange. Yet his awards and publications tell only part of the story. The deeper challenge lies in what happens between the two languages.
Egyptian translator Yahya Mokhtar, left, meets renowned Chinese writer Alai. They are pictured with Alai’s novel The Dust Settles. (Photo/Ayman El-Kady, Bridging News Cairo Bureau)
Choosing the right story for the Arabic reader
For an Arabic translator, Chinese is difficult for reasons that go far beyond its thousands of characters and tonal pronunciation.
Mokhtar says the greater challenge lies in the history, culture and meaning embedded in the language. Proverbs, idioms and cultural references often cannot be translated word for word into Arabic.
“There are many people who master a language but do not master translation,” he said.
For Mokhtar, translation is an art built through study and practice. Translators must work across journalistic, political, scientific, economic, legal and literary texts before reaching a professional level.
Literature is especially demanding. The translator must carry more than information between languages, preserving the original work’s atmosphere, rhythm, cultural references and emotional meaning.
Mokhtar tends to avoid works with highly complex historical and cultural backgrounds. Heavy with local references, historical events and ancient philosophies, they can be difficult for readers unfamiliar with that context. Adding too many explanatory notes, he says, can weaken the spirit of literary translation.
He prefers short novels and novellas, describing them as complete, satisfying meals that are easier and quicker to digest. There is also a practical reason: Chinese works often become significantly longer in Arabic, making an already lengthy novel even more daunting.
For Mokhtar, a translator must consider not only the original author, but also the reader discovering that author in another language.
A published profile of Egyptian translator Yahya Mokhtar presents his work promoting literary exchange between China and the Arab world, alongside several Chinese books he has translated into Arabic. (Photo/Ayman El-Kady, Bridging News Cairo Bureau)
From translator to cultural bridge
A finished translation begins with work that closely resembles a writer’s. Before starting a project, Mokhtar researches the author’s life, style and reputation, reads reviews and academic studies, and, most importantly, finishes the entire book.
Then he prepares his tools: a computer, dictionaries, reference materials and a quiet workspace. Unlike many translators, he starts early.
“I like to work in the early morning,” he said.
Mokhtar also sets a daily target and makes sure he meets it. The routine underscores a central truth about translation: language skills alone are not enough. Recreating a book requires discipline and a deep understanding of the original text.
Mokhtar’s work now extends beyond translation. He researches Chinese literature, teaches students in Egypt and participates in exchanges between Egyptian and Chinese writers and intellectuals.
He sees this work as essential to bilateral relations. Political and economic ties may provide the framework, but cultural exchange builds familiarity between people. For him, success is measured not only by agreements and investments, but also by how each society views the other.
His journey reflects that idea. He began as an Egyptian student learning Chinese, spent seven years living and studying in China, and returned home to bring Chinese literature to Arabic readers. He has also taught Arabic in China and joined exchanges aimed at bringing the two cultures closer.
His career shows that cultural bridges are often built through the patient work of individuals who learn another language deeply enough to understand the world behind it.
Chinese was once simply Mokhtar’s university subject. Today, it is the foundation of his career—and a bridge between two civilizations."
Vivian Yan
https://www.ichongqing.info/2026/09/02/he-chose-to-learn-chinese-seven-years-later-hes-bridging-egypt-and-china/amp/
#metaglossia
#metaglossia_mundus
Traduction d'extraits de naissance en ligne
"Le ministère de l’Intérieur a annoncé, le 31 août 2026, qu’il est désormais possible de demander la traduction d’un extrait de naissance en ligne.
Selon le ministère de l’Intérieur, ce service est disponible sur e-bawaba.tn . L’accès au site web nécessite la saisie de l’identifiant numérique mobile E-Houwiya.
La demande de traduction sera, automatiquement, communiquée à la municipalité où la naissance a été initialement enregistrée.
Ce service permettra, ainsi, de profiter de la traduction de l’extrait de naissance sans se rendre auprès des autorités concernées."
https://espacemanager.com/ministere-de-linterieur-la-traduction-de-lextrait-de-naissance-disponible-en-ligne/
#metaglossia
#metaglossia_mundus
"Le Vatican a accueilli l'une des journées de la troisième conférence du Centre international d'études de la conscience (ICCS), consacrée à l'impact des technologies sur l'art, la culture et l'esprit humain, ce 2 septembre 2026.
Au Vatican, philosophes et scientifiques interrogent les relations entre homme et IA
La Casina Pio IV, dans les jardins du Vatican, accueille l'une des rencontres de la troisième conférence du Centre international d'études de la conscience (ICCS). Intitulée Créativité: Esprits et Machines, la journée a été consacrée à l'impact des technologies sur l'art, la culture et l'esprit humain. Selon le fondateur de l'ICCS, si la créativité de l'IA explose, «l'intelligence artificielle est comme une voiture rapide: il faut faire attention à ne pas percuter un mur».
Eugenio Murrali - Cité du Vatican
Pour Dmitry Volkov, l'intelligence artificielle est comme une voiture: elle nous permet une accélération extraordinaire, mais lorsqu'on conçoit une belle voiture, surtout une voiture de course, «il faut accorder une grande importance à la conception des freins, très efficaces et fiables, pour éviter qu'elle ne s'écrase contre un mur». Le philosophe et entrepreneur technologique, fondateur du Centre international d'études de la conscience (ICCS), résume par cette image les possibilités offertes par cette technologie qui transforme notre époque et que la troisième conférence annuelle de l'ICCS explore en profondeur, soulevant les questions, parfois dérangeantes, nécessaires à l'analyse de l'impact de l'IA sur l'humanité.
L'IA est-elle vraiment créative ?
Lancée hier à l'Université Roma Tre dans la capitale italienne, la conférence ICCS –Créativité: Esprits et Machines– s'est poursuivit ce mercredi à la Casina Pio IV au Vatican, avant de se conclure ce 3 septembre. Cette conférence réunit des philosophes, des neuroscientifiques et des experts internationaux en intelligence artificielle afin de débattre de la capacité de l'IA à être véritablement créative. Pour l'instant, selon Dmitry Volkov, une chose est sûre: l'IA peut contribuer à développer la créativité humaine, mais avec la nécessité d'être vigilants et de mettre en place des mécanismes de contrôle.
Le philosophe italien Pietro Perconti, cofondateur de l'ICCS, estime également qu'«il n'est pas absurde de parler de créativité artificielle». «Bien entendu, de nombreuses mises en garde s'imposent ; la prudence est de mise. La créativité issue des machines diffère de celle que nous avons l'habitude de concevoir. Il s'agit d'une créativité différente, à apprécier comme une entreprise collective plutôt que comme le fruit du génie d'un seul artiste, comme on a généralement tendance à le penser en Occident.»
Les vidéos sont un parfait exemple des avantages et dangers de la créativité des machines. Dmitry Volkov se souvient que «pendant de nombreuses années, produire une vidéo, tourner un film ou créer un dessin animé était une activité réservée à une poignée de personnes». En effet, cela exigeait d'apprendre à dessiner et à animer des images sur ordinateur, tandis que l'IA a désormais permis à des milliers de personnes de créer leurs propres films ou dessins animés – bref, de s'exprimer. Le philosophe met toutefois en garde contre un effet secondaire de ce qu'il appelle une «explosion de créativité»: les deepfakes , la manipulation de photos, de vidéos ou d'images qui nuit aux individus.
«L'effet sycophante»
L'intelligence artificielle a tendance à faire en sorte de plaire à ses utilisateurs. Cette disposition est qualifiée d'«effet sycophante», un biais de complaisance, qui va jusqu'à flatter l'usager au détriment de l'exactitude des informations. Ce comportement, avertit Dmitry Volkov, «conduit parfois une personne à être récompensée de son narcissisme et d'idées insuffisamment fondées. C'est une caractéristique importante dont nous devons être conscients.» Cette attitude des machines peut être atténuée: «les nouveaux modèles actuellement disponibles sont en réalité beaucoup plus indépendants et autonomes.»
Lire aussi
Léon XIV appelle à humaniser la technologie au service de la vie et de la paix
28/08/2026
Léon XIV appelle à humaniser la technologie au service de la vie et de la paix
À ce stade, il est légitime de se demander si l'IA est ou pourrait devenir consciente. La position du fondateur du centre de réflexion est nuancée: «En philosophie, nous avons un terme technique précis pour la conscience: qualia.» Qualia, terme latin dérivé de quale, désigne les aspects qualitatifs des expériences conscientes. Ils correspondent à «une sensation particulière, un sentiment subjectif, quelque chose que l'on ressent à l'instant présent».
Par exemple, explique-t-il: «si je vois du rouge, j'ai l'impression de voir du rouge, donc j'en suis convaincu; c'est l'un des qualia les plus simples et fondamentaux.» Poursuivant son explication, le philosophe parvient à cette conclusion personnelle: «Je pense que la recherche n'en est qu'à ses débuts. Mais d'une manière générale, je crois que les humains sont des systèmes mécaniques et que nous verrons des phénomènes similaires chez les ordinateurs. Ainsi, à un moment donné, nous serons capables de repérer une IA consciente et nous devrons apprendre à la gérer.»
Apprendre à intéragir avec l'IA
Durant cette journée au Vatican, comme l'explique le philosophe italien, les discussions poursuivent un objectif ambitieux: «tenter de comprendre s'il est possible, voire pertinent, d'imaginer des relations entre êtres humains et compagnons artificiels – amitiés, relations d'attention, relations sentimentales.» Les intellectuels réunis au Vatican s'interrogent sur la façon dont ces relations peuvent être productives et, au contraire, préoccupantes. Dans leurs échanges, les experts gardent à l'esprit le message de l'encyclique Magnifica Humanitas de Léon XIV: «Ce que dit l'encyclique, à savoir placer l'humain au centre, n'est pas seulement un besoin religieux, c'est un besoin humain.»
Un message qui a guidé les participants de la conférence. «Étant donné que notre monde se transforme en un monde de relations numériques, comment pouvons-nous prendre en compte cette préoccupation humaniste fondamentale dans un monde qui devient lui aussi numérique ?» Pour le philosophe, il existe une réponse: «Nous devons nous éduquer aux relations avec l'intelligence artificielle, et il s'agit essentiellement d'entreprendre une phase d'éducation publique.»"
Au Vatican, philosophes et scientifiques interrogent les relations entre homme et IA - Vatican News https://share.google/HvIMm5JqLscFjvv20
#metaglossia #metaglossia_mundus
"Tether Releases Open-Source AI Translation Models for African and European Languages
New models run privately and offline on everyday devices, with TranslatePsy-AfriSLM outperforming far larger systems, opening a path to local-language education and health knowledge for underserved communities across Africa; a parallel TranslatePsy-EuroNano release brings the same technology to nine European languages
2 September 2026 – Tether AI Research today announced the launch of new families of open-source AI translation models designed to run directly on smartphones, laptops, and other edge devices with a primary focus on Sub-Saharan Africa. The release includes QVAC TranslatePsy-AfriSLM and QVAC TranslatePsy-AfriNano, supporting 19 and 8 African languages respectively, alongside QVAC TranslatePsy-EuroNano covering 9 European languages.
By processing translations locally, the TranslatePsy models can operate without an internet connection while keeping users’ data on their devices rather than sending it to third-party cloud servers.
What Translation Unlocks For The Underserved
For hundreds of millions of people across Africa, one of the barriers to modern AI is language. The most powerful tools run in a few major languages and depend on the cloud, putting them out of reach for many.
TranslatePsy-AfriSLM supports Hausa, Amharic, Yoruba, Lingala, Swahili, Igbo, Zulu, Somali, Oromo, Malagasy, Kinyarwanda, Xhosa, Afrikaans, Wolof, Luganda, Nyanja, Shona, Tswana, and Southern Sotho. Together, these languages span West, East, Central, and Southern Africa and represent roughly half of Africa’s population.
Translation across 19 African languages can unlock the ability to deliver courses, educational content, scientific material, and AI-powered learning tools directly to children and adults in their own languages.
Despite containing just 800 million parameters, the smallest TranslatePsy-AfriSLM model outperformed Qwen3.5-122B-A10B, TranslateGemma-27B, and NLLB-3.3B across the FLORES-200, BOUQuET, and SMOL translation benchmarks.
A key innovation is the introduction of a new quality-estimation filtering method that removes up to 96% of low-quality open-source training data. By improving the quality of the underlying data, Tether AI Research was able to achieve stronger translation performance with significantly smaller models.
Healthcare will be one of the highest-impact applications. Patients may speak different local languages, while connectivity can be unreliable in the communities that need information most.
Combined with Tether QVAC MedPsy, a small foundation model for medical and healthcare applications, TranslatePsy-AfriSLM creates a potential pathway to deliver medical knowledge and health education in the local languages of hundreds of millions of people.
Such systems would require appropriate safeguards and clear boundaries between health education and clinical care, but the potential impact is substantial.
Agriculture, Humanitarian Response, and Cross-Border Communication
The applications extend well beyond education and healthcare. Farmers could receive agricultural information in local languages, helping translate technical knowledge into practical guidance. In humanitarian and disaster-response settings, where camps and affected areas may have poor connectivity.
For NGOs and field organizations, local-language translation could help field workers communicate across multiple communities without carrying separate translation systems for every language.
Tether has spent years building physical touchpoints in these communities. Across Sub-Saharan Africa, its solar-powered kiosks let residents charge a phone, swap a battery, and access digital financial services where the grid and the banking system do not reach. Those same hubs could become places where a family charges a phone by day and, by evening, where children watch a science documentary in their own language, or parents learn new farming techniques from a local-language video, turning a charging point into a point of access to knowledge.
The Same Approach, Applied in Europe
The same design principle underpins a parallel release for European languages. Tether AI Research’s TranslatePsy-EuroNano replaces dozens of separate bilingual models with two compact multilingual models per performance tier. Using English as a pivot, the models support 90 translation directions across nine European languages. The smallest deployment requires just 36MB of storage, compared with 633MB for an equivalent Firefox offline translation configuration, reducing storage requirements by approximately 94%.
TranslatePsy-EuroNano, the highest-quality model in the European family, retained 98.4% of Meta’s NLLB-200 translation quality when translating into English while using a fraction of the storage required by larger systems.
“Four billion people were left behind by the traditional financial system, and the most powerful technology of our age has repeated that failure,” said Paolo Ardoino, CEO of Tether. “Language should not determine who can benefit from artificial intelligence. Open translation models like these are a step toward a future where education and AI tools reach hundreds of millions of people who have neither reliable connectivity nor access to expensive systems. A mother could get real medical information she understands, instead of guessing. A child could learn in their own language. That is the future we are building through QVAC.”
TranslatePsy-AfriSLM is available for download on Hugging Face at this link, in three sizes (full-precision and smaller quantized versions):
qvac/TranslatePsy-AfriSLM-0.8B
qvac/TranslatePsy-AfriSLM-2B
qvac/TranslatePsy-AfriSLM-4B
TranslatePsy-Nano is available for download on Hugging Face at this link: https://huggingface.co/collections/qvac/translatepsy-nano. It supports both European and African language translation, with models offered in full-precision and quantized versions: :
qvac/TranslatePsy-EuroNano
qvac/TranslatePsy-AfriNano
The research underpinning TranslatePsy-AfriSLM has also been accepted for presentation at the Empirical Methods in Natural Language Processing (EMNLP) 2026 conference.
About Tether AI Research
Tether AI Research is part of Tether’s broader vision to advance freedom, transparency, and innovation through technology. Its mission is to enable people and organizations to connect and share information directly, without unnecessary intermediaries. By creating secure, peer-to-peer systems, Tether AI Research gives users greater control over their data, communications, and digital interactions. Tether AI Research aims to redefine how information flows across networks by replacing centralized models with decentralized infrastructure designed for privacy, efficiency, and resilience.
*References to Tether AI Research mean Tether Data, S.A. de C.V.
About QVAC
QVAC is Tether’s advanced AI research initiative dedicated to building open, decentralized, and adaptive intelligence systems. Its mission is Local AI and Infinite Intelligence. It is guided by an uncompromising vision of a world where AI lives and learns on any device, empowering individuals and communities rather than concentrating power in corporate data centers."
https://tether.io/news/tether-releases-open-source-ai-translation-models-for-african-and-european-languages/
#metaglossia
#metaglossia_mundus
"L'ouvrage d’Issa Hassane sur les enfances en crise traduit en cinq langues
Publié en français en 2025 aux Éditions Universitaires Européennes, l’ouvrage du Dr Issa Hassane consacré à la représentation de l’enfance dans trois œuvres majeures de la littérature africaine est désormais disponible en anglais, en allemand, en italien, en espagnol et en portugais.
Il y a un an, le 4 juin 2025, nous annoncions la publication de deux ouvrages d’Issa Hassane aux Éditions Universitaires Européennes : Enfances en crise dans la littérature africaine : L’Aventure ambiguë de Cheikh Hamidou Kane, Allah n’est pas obligé d’Ahmadou Kourouma et Le Chercheur d’Afriques de Henri Lopes et Imaginaires du voyage et de la migration dans les cultures sahéliennes : textes-écrans d’Issa Hassane.
Un an après cette première parution, le premier ouvrage est désormais disponible dans cinq nouvelles langues : l’anglais, l’allemand, l’italien, l’espagnol et le portugais.
Il paraît en anglais sous le titre Children in Crisis in African Literature, en allemand sous celui de Krisenhafte Kindheiten in der afrikanischen Literatur, en italien sous le titre Bambini in crisi nella letteratura africana, en espagnol sous celui de Niños en crisis en la literatura africana et en portugais sous le titre Crianças em crise na literatura africana.
À travers cette étude, Issa Hassane analyse la représentation de l’enfance en situation de crise dans trois œuvres importantes de la littérature africaine : L’Aventure ambiguë de Cheikh Hamidou Kane, Allah n’est pas obligé d’Ahmadou Kourouma et Le Chercheur d’Afriques de Henri Lopes.
La traduction de cet ouvrage en cinq langues élargit son accessibilité et permet à de nouveaux lecteurs de découvrir cette réflexion consacrée à l’enfance, à ses épreuves et à sa place dans les imaginaires littéraires africains.
Les différentes versions sont proposées sur plusieurs plateformes de vente en ligne, notamment Amazon et Momox Shop."
Rédigé le 30 Août 2026. Publié dans Culture.
Source : informations communiquées à Nigerdiaspora par Dr Issa Hassane.
Boubé G. (Nigerdiaspora)
https://nigerdiaspora.net/culture/litterature-louvrage-dissa-hassane-sur-les-enfances-en-crise-traduit-en-cinq-langues
#metaglossia #metaglossia_mundus
Center for Assessment of Sign Language Interpretation (CASLI)
"NEW CASLI Exam Rates
Effective October 1, 2026
Please be advised that CASLI examination fees will increase effective October 1, 2026.
Examination fees have not been adjusted since January 2023. This adjustment is necessary to reflect the continued increase in the costs associated with administering, maintaining, and supporting CASLI’s examination programs.
The updated fees will help CASLI maintain the quality, security, and accessibility of its examination programs while providing the resources needed to sustain and strengthen CASLI’s testing services for the future. This includes supporting ongoing program improvements, technology and infrastructure needs, examination development and maintenance, as well as opportunities to expand and grow CASLI’s testing program for RID certification program to meet the evolving needs of the profession.
Important Fee Information
Examinations requested on or before September 30, 2026 will retain the current examination fee during the 30-day purchase period.
Examinations requested on or after October 1, 2026 will be subject to the new examination fees.
To request an exam, log into your CASLI account and select Request an Exam to complete and submit the Exam Request Form.
We recognize that the increase in examination fees may present financial challenges for some candidates. Candidates are encouraged to explore available financial assistance opportunities, including the RID Scholarship Program, RID state-affiliated chapters, state and local organizations, and interpreting agencies that may have funds or other resources available to help support the cost of examinations.
We appreciate your understanding and continued support of CASLI as we work to strengthen our certification programs and build a sustainable foundation for future growth and opportunities.
For complete information regarding the updated examination fees, please visit our website.
For the new exam rates, please go to: New CASLI Exam Rates as of October 1, 2026"
https://www.casli.org/2026/08/27/new-casli-exam-rates/
#metaglossia
#metaglossia_mundus
"Hayy Jameel Library research lab explores translation beyond language
special
A research lab at Jameel Library is inviting people to take a broader view of translation — as a way of moving between languages, images, sounds, experiences and different forms of knowledge.
Short Url
https://arab.news/gfa6y
The Task of the Translator Research Lab, developed and facilitated by artist and editor Sarasija Subramanian, runs until Aug. 30
Brings participants together for collective reading, watching and listening sessions, alongside exercises exploring language, accessibility, subjectivity and the idea of the original
Updated 31 August 2026 16:22
JEDDAH: A research lab at Jameel Library is inviting people to take a broader view of translation — as a way of moving between languages, images, sounds, experiences and different forms of knowledge.
The Task of the Translator Research Lab, developed and facilitated by artist and editor Sarasija Subramanian, runs until Aug. 30. It brings participants together for collective reading, watching and listening sessions, alongside exercises exploring language, accessibility, subjectivity and the idea of the “original.”
“The starting point for us was the idea of translation as something much bigger than moving from one language to another,” said Khaled Al-Qahtani, library and research coordinator at Jameel Library.
“We are constantly translating between languages, but also between images and words, between what we see and what someone else might experience, between different ways of knowing and understanding.”
The week-long program explores what happens when an image is described for someone who cannot see it, when sound is translated into words, or when an experience is communicated through another form.
Al-Qahtani said: “We wanted the lab to create a space where we could slow down and actually examine those processes. What happens when you describe an image for someone who cannot see it? What happens when you translate sound into words, or an experience into another form? What is gained, what is lost, and what kinds of new meanings can emerge?”
For Subramanian, the project comes from her interest in “expanded translation,” a concept that considers movement between mediums as translation.
“The lab comes from my own interest as an artist and editor in the field of expanded translation, where translation is not only about moving from one language to another, but also thinking of information and ideas moving from one medium to another,” she said.
Subramanian works with drawing and tracing as forms of translating images, while her publishing house, Reliable Copy, works with artists to translate their ideas and research into books.
“The primary drive of the lab is to be able to consider these kinds of shifts between mediums as translation, and consequently to question translation’s need to be faithful to the ‘original’, and to rethink the mandate that we must be objective and impersonal during the process,” she said.
The lab also examines accessibility through tools including alt-text, audio descriptions and closed captions.
Subramanian said such tools were often viewed as additional accommodations for visually impaired audiences but could also encourage people with sight to look more closely.
“When we write alt-text for a painting, or audio-descriptions for a film, we see the image or the video through a very different point of view, which reveals to us details and nuances that we would otherwise overlook,” she said.
She compared translating an image to translating a text from English into Arabic, where every word, punctuation mark and piece of context must be considered.
“It is the same with an image, or a video,” she said, adding: “As such, the lab as a whole is to look at material we encounter on a daily basis closely, and rethink the ways in which we communicate and share.”
Subramanian said this approach also challenges the idea of a single, fixed interpretation, allowing different perspectives to become part of the process: “The question then, or the task, is to be able to use these subjectivities as a tool and an aid, instead of seeing them as inherently wrong.”
For Al-Qahtani, Jameel Library was a natural setting for the project because libraries already bring together different languages, histories and forms of knowledge.
“Libraries are already spaces of translation in a way,” he said.
The project, he added, sought to make the library more than a place for presenting research, turning it into a space where research could unfold collectively through experimentation, conversation and making.
Accessibility also requires cultural institutions to question assumptions about how audiences are expected to look, listen, read or participate.
“We often design cultural experiences around an assumed audience and an assumed way of looking, listening, reading or participating,” Al-Qahtani said. “Accessibility asks us to question those assumptions.
“For me, accessibility is not something that should come at the end of a process, as an additional service or adjustment. It can actually change the way we think about the work from the beginning.”
Guest lecturer Sara H. Hammami, a librarian, book artist and poet, sees translation as closely connected to culture and human connection.
She said translation could create movement toward understanding, curiosity and respect, but that this required openness from everyone in a multilingual environment.
“In any multilingual space this requires a curiosity regardless of the language one is starting in/moving towards it must be reciprocated by everybody for the kind of fluidity and warmth that can be brought forward from a space of translation,” she said.
Her work across libraries and translation has also shaped her view of accessibility. Cultural institutions, she said, have a responsibility to make audiences feel welcome: “The more people are given the opportunity and chance to attend and participate in these conversations and spaces, the stronger and more imaginative we become as a cultural landscape,” Hammami said.
In Saudi Arabia, she sees language equity as an important starting point.
“I feel language equity in a multilingual space is one of the first and most accessible ways to start building accessibility for our public,” she said. “I wonder what it might do for us and our language if we were to enter more cultural landscapes and find materials not only in Arabic, English, or French, but also in Tagalog, Urdu, or Bengali. I imagine a future in which everyone feels invited to these spaces and moments.”
Artist Park McArthur also approaches translation through questions of access, language and power.
“Translation and access could be shared phenomenon known by different names,” McArthur said, adding that stories and ideas move between languages through translation, whether through curiosity or other forces: “The power dynamics of language and access make something like force hard to pin down, actually. Power is an element of language and access.”
McArthur said language also extends beyond the linguistic.
“Artistic traditions and experimentation teach me that language is linguistic, while it is also a way of doing things, a means of gathering, a ritual, a performance,” she said.
She wants audiences to reconsider the traditional relationship between an original work and its translation. She also argued that the original also depends on translation for its circulation and continued use.
“We are perhaps more accustomed to approaching the act of translation as ‘dependent upon’ an originary text. An originary text is dependent upon versioning and translation for its circulation, its use, for its ongoing care,” she said.
Her own contribution to the program illustrates that relationship. An artistic experiment describing a day at the beach as a space for gathering was translated from English into Arabic by Al-Qahtani.
“Khaled’s translation, which I have no linguistic access to, extends this gathering, this day at the beach, which I dream of,” McArthur said.
While the lab is designed for participants who responded to the open call, its wider public programming opens the conversation to a broader audience."
Afshan Aziz
August 31, 2026 16:08
https://www.arabnews.com/saudi-arabia/hayy-jameel-library-research-lab-explores-translation-beyond-language-2656443
#metaglossia
#metaglossia_mundus
"
500 Years of the Tyndale Bible
Five centuries later, Tyndale’s work still shapes our Bibles, language, and theology.
09/1/26
John Stonestreet
Glenn Sunshine
Among the many milestones marked in 2026 is the 500th anniversary of William Tyndale’s translation of the New Testament, and later the Old Testament, into English. His commitment that common people should have the Bible in their own language eventually led to his execution. Still, his translation has influenced English Bibles ever since.
Tyndale, an Oxford scholar, was ordained to the priesthood in 1515. A reformer in the church, he was heavily influenced by Martin Luther. In fact, it was because Luther published a translation of the New Testament into German in 1523 that Tyndale decided to do the same for English.
Two centuries prior, John Wycliffe had been condemned by the Catholic Church as a heretic for translating the Bible into Middle English. Because every unauthorized translation of the Bible was prohibited as heresy, Tyndale’s bishop refused to grant permission and likely flagged Tyndale as a Lutheran sympathizer. Since Henry VIII was executing Lutherans for heresy, this became very dangerous for Tyndale.
So, in 1524, Tyndale fled to Germany so that he could translate the Scriptures, print them, and smuggle copies back to England. He sent the first copies of the New Testament to England in 1526 and then began work on the Old Testament. His New Testament was quickly condemned by Bishop Tunstall of London, who warned booksellers against it, bought all available copies, and publicly burned them. His actions, however, did not have their desired result. Even many faithful Catholics questioned whether it was appropriate for an executioner to burn the Bible.
Henry VIII appealed to Holy Roman Emperor Charles V to have Tyndale extradited to England on charges of heresy. Though Charles was no friend to Lutherans, he refused on the grounds that Henry had presented no formal evidence. In 1530, Tyndale further angered Henry by publishing The Practice of Prelates. In it, he opposed Henry’s annulment of Catherine of Aragon (the first of the king’s wives to be annulled) and argued against the misbehavior of Catholic clerics.
In 1535, Tyndale was arrested in Antwerp on suspicion of heresy and imprisoned in Brussels. Rather than stand trial, the Roman Catholic inquisitor Jacobus Latomus allowed Tyndale to defend his theological positions in written communications. The two debated with replies and counter-replies, from which Latomus wrote a book. Interestingly, Bible translation was not an issue in these discussions. In fact, Tyndale was allowed to continue translating the Old Testament while in prison.
Latomus concluded that despite several important points of agreement between the two, Tyndale held views common to Protestants that had been condemned by the Catholic Church. Though Latomus seems to have attempted to save Tyndale, he was ultimately unsuccessful. Tyndale’s ideas were found heretical according to Rome, a capital offense.
On October 6, 1536, William Tyndale was tied to a stake, strangled, and his body burned. He was 42.
Though he never completed his translation of the entire Bible, his legacy and influence continue. His translation of the entire New Testament, the Pentateuch, and Jonah were used in the Coverdale Bible, published a year before his execution and the first complete Bible published in Early Modern English. The Matthew Bible of 1537 and the Great Bible of 1539, which was the first royally authorized English Bible, also used his translation of these books. In fact, a majority of the New Testament and the first half of the Old Testament in the King James Version were likely taken from Tyndale’s translation.
Tyndale coined common words and phrases used in translations such as Passover, scapegoat, mercy seat, and atonement—a word with the etymological breakdown of “At one with God.” He also coined phrases such as “Let there be light,” “the powers that be,” “it came to pass,” and “the signs of the times.”
Tyndale’s work laid the foundation for virtually every English Bible translation of the past 500 years and left a major mark on the English language and our theological vocabulary. The 500th anniversary of his New Testament is an opportunity to be grateful for his work and its impact on the world."
https://breakpoint.org/500-years-of-the-tyndale-bible/
#metaglossia #metaglossia_mundus
"by: Ben Breen
Posted: Sep 1, 2026 / 03:50 PM EDT
Updated: Sep 1, 2026 / 03:50 PM EDTVERMON (ABC22/FOX44) – Earlier this year, a complaint alleged Vermont prisons have failed to fairly accommodate ICE and Border Patrol detainees who don’t speak English. The Department of Corrections has now agreed to provide translation and language resources for all prisoners.
The complaint, back in January, was filed by the Vermont Human Rights Commission – and accused the Vermont DOC of restricting access to attorneys, not using qualified interpreters, and failing to provide translations for important instructions – like on how to seek medical care.
According to the agreement, Vermont prisons will have to determine the language in which each inmate is most fluent, and arrange for translations. Vermont Asylum Assistance Project staff, including attorneys, will also be allowed to confidentially meet with anyone detained on behalf of ICE or Border Patrol every two weeks.
Read the full 29-page agreement hereDownload
The agreement also requires information on seeking medical care to be posted in the 10 most common languages spoken in Vermont.
As of U.S. Census data released in 2025, that would mean English, French, Spanish, German, Nepali, Chinese, Bosnian, Italian, Russian, and Mandarin Chinese.
CSWD partners with VLJP to create ‘How-To Recycle Right’ Videos in 15 Languages
“This agreement ensures that all people incarcerated in Vermont, including ICE and Border Patrol detainees, will not be denied equal treatment in state correctional facilities due to language barriers,” says Big Hartman, Executive Director of the HRC. “We appreciate DOC’s willingness to … discuss practical measures to provide equal treatment for individuals housed in DOC facilities who do not speak fluent English.”"
by: Ben Breen
Posted: Sep 1, 2026 / 03:50 PM EDT
Updated: Sep 1, 2026 / 03:50 PM EDTVERMONT
https://www.mychamplainvalley.com/news/local-news/vermont/vermont-inmates-to-be-provided-translation-language-resources/
#metaglossia
#metaglossia_mundus
Luma aims to help Arabic teams in the region to develop and deliver creative work in a more natural and intuitive way.A creative intelligence platform in Arabic designed for creative communities
Luma Agents currently supports Arabic-language prompting and generates Arabic-language creative outputs
Luma aims to help Arabic teams in the region to develop and deliver creative work in a more natural and intuitive way.
"Luma Agents currently supports Arabic-language prompting and generates Arabic-language creative outputs. The new update will localise the total user experience, including navigation, menus and workflow tools.
Jason Day, Head of EMEA at Luma AI, announces the Arabic interface during a panel at LEAP 2026 in Riyadh, Saudi Arabia.
Luma AI has revealed that Luma Agents, its unified AI creative platform, will offer a fully native Arabic-language interface later this year.
This will be a creative intelligence platform in Arabic designed for creative communities in Saudi Arabia and the wider MENA region.
Luma Agents currently supports Arabic-language prompting and generates Arabic-language creative outputs. The new update, announced at LEAP 2026 in Riyadh, will localise the total user experience, including navigation, menus, and workflow tools.
The update is a direct result of Luma’s strategic partnership with HUMAIN, as well as engagement with creatives and enterprises in Saudi Arabia.
Arabic will be the first non-English interface language available on Luma Agents, enabling creators across Saudi Arabia and the wider Arabic-speaking world to work in their native language as they move between ideas and production workflows.
Luma AI expands access to Arabic creative AI
Arabic-language interfaces remain uncommon among AI generation platforms, despite rapid growth in demand for AI-powered content creation across the GCC and wider MENA region.
By localising the Luma Agents interface, Luma aims to remove a practical barrier for creative professionals working primarily in Arabic, allowing teams in the region to develop and deliver creative work in a more natural and intuitive way on a platform that speaks to their language.
It also forms part of Luma’s broader investment in the country – including the establishment of its regional headquarters in Riyadh.
“Over the past year, we have been listening to and understanding the needs of the creative community in Saudi Arabia,” said Jason Day, Head of EMEA at Luma AI.
Day added, “The process of localizing the interface has been guided by the creative community, for the creative community in the Kingdom. Providing Luma Agents in Arabic enables the 400 million Arabic speakers worldwide to be prolific – going from first idea through to the finished outcome without the friction of working in multiple languages.”
The fully localised Arabic-language interface for Luma Agents will begin rolling out later this year."
Sep. 02, 2026
Anup Oommen
https://campaignme.com/luma-ai-to
#metaglossia
#metaglossia_mundus
"Lost in translation: Edmonton case reveals longstanding worries about unqualified interpreters in Alberta courts
"This type of situation may lead to unnecessary adjournments, appeals and new trials," the report states
Hundreds or thousands of people facing criminal charges in Alberta may have received unfair trials due to subpar language translation services, lawyers say in response to recently disclosed documents detailing problems with the province's court interpreter system.
Hundreds or thousands of people facing criminal charges in Alberta may have received unfair trials due to subpar language translation services, lawyers say in response to recently disclosed documents detailing problems with the province’s court interpreter system.
An Edmonton judge is currently hearing arguments in the case of Ramjit Sandhu, who was convicted earlier this year of sexual assault. The conviction comes after Sandhu’s second trial, after an earlier guilty verdict was thrown out over “seriously inadequate” translation by the Punjabi language interpreter appointed to the case.
Sandhu’s lawyer, Kristofer Advent, has since obtained internal court documents showing both courts and government knew about longstanding problems with how Alberta courts handle language interpreting.
One document, produced in 2021 by a now-defunct review committee, said “the hiring of unqualified interpreters is an issue” and not “isolated incident(s).”
“This type of situation may lead to unnecessary adjournments, appeals and new trials,” the report states. The committee ultimately opted against making any recommendations, though, because members were “unable to reach a consensus with respect to the government’s involvement in funding, issues of training and accreditation requirements.”
Advent said he was shocked to read the report, which he received this past spring as part of an evidence disclosure application. While many lawyers have stories about inaccurate courtroom translation, the defence bar and the public were never informed of the systemic issues courts and government knew about, he said.
“Basically what they acknowledge in there is ‘inaccurate resolution of cases,’ which is nice language for wrongful convictions and miscarriages of justice,” Advent said. “As far as I’m aware, nothing has been done about it."
By Jonny Wakefield
Published Aug 31, 2026
https://edmontonjournal.com/news/crime/lost-in-translation-edmonton-case-reveals-longstanding-worries-about-unqualified-interpreters-in-alberta-courts
#metaglossia
#metaglossia_mundus
"‘Strong suspicion of bad faith’: Interpreters still fighting for lost wages after government breached contract
The federal procurement watchdog recommended Public Services and Procurement Canada pay damages to four complainants for lost work after the department violated contracts with freelance interpreters.
When the work suddenly dried up, Pamela Cousineau thought about leaving the country.
After three decades as a freelance interpreter, she was accustomed to setting money aside for just such a downturn.
But the spring of last year was different.
“It happened overnight without any prior warning — and it was bad,” she said. “People were anxious. Everybody was anxious.”
After Justin Trudeau resigned as Liberal leader in January 2025, Gov. Gen. Mary Simon agreed to prorogue Parliament while the party searched for his replacement.
For freelance interpreters — who provide live translation services at public meetings on Parliament Hill and for federal departments — the pause on Parliamentary business meant the sudden cancellation of months of work.
“We thought when (the federal Translation Bureau) had cancelled, that meant they had no work,” Cousineau said. “It’s okay if that’s the case. The problem was that wasn’t the case.”
As it turned out, the work hadn’t stopped for everyone.
Instead, according to four interpreters who filed complaints with the federal procurement watchdog, the government breached the terms of the interpreters’ contracts “in secrecy” by awarding the available work to a subset of freelancers who were charging the lowest prices.
In a series of reports published earlier this year, the Office of the Procurement Ombudsman agreed that in doing so the government had violated its contracts with the interpreters. The ombud recommended Public Services and Procurement Canada (PSPC) pay out damages for the lost wages.
In each case, the office described “a strong suspicion of bad faith” on the part of government, but ultimately decided the evidence wasn’t strong enough to judge intent.
Feds finalize new rules for freelance interpreters, risking shortages on Parliament Hill
Could new rules worsen an interpreter shortage in Parliament?
About six months later, Cousineau says she and the other complainants are still fighting for the wages they say they’re owed.
‘Carried out in secrecy’
According to a timeline laid out in the reports, three freelance interpreters sent letters to PSPC on April 25, 2025, expressing concerns over a “significant deviation” from the terms of their contracts.
Under those contracts, the government was required to award work to qualifying interpreters at random once it had determined enough suppliers met the requirements, including quality, for a given event.
Through conversations with colleagues, Cousineau got a sense those rules weren’t being followed.
“We know each other,” she said. “There aren’t many of us.”
Among the pool of about 70 freelance interpreters working with the federal Translation Bureau, Cousineau said all the work was going to just three interpreters, while the rest were getting nothing.
This advertisement has not loaded yet, but your article continues below.
Article content
“Some of them were actually complaining that they were getting so much work that they didn’t have a single day off,” she said.
So, she wrote to the bureau, seeking an explanation.
After hearing nothing for weeks, one of the four complainants followed up on May 12, according to the ombud. In June, another followed up again.
They were met with silence until June 17, when PSPC promised individual answers to all the interpreters.
“It took a very, very long time,” Cousineau said. “It took about three months before we got a letter back from the government saying, ‘Yeah, you’re right. We haven’t been following the rules.’”
According to the ombud reports, PSPC admitted it “did not fully” respect the contract between Feb. 1 and April 28, 2025, saying the breach was due to a “misunderstanding” of one clause as well as “internal miscommunication.”
In a Feb. 24 response to the reports, PSPC said it “promptly conducted an internal review” after receiving the initial letter in April.
“Additional steps have already been taken to ensure compliance with contract clauses, such as centralizing the work attribution function under a single authority to ensure better oversight,” the statement read.
The interpreters were unconvinced.
“The pattern of conduct in the administration of this contract — as evidenced in this sworn statement — leaves no doubt: the violation was not accidental, but the product of intent to deceive, bad faith, and knowing and intentional wrongdoing — a deliberate course of action carried out in secrecy,” one wrote in their complaint.
In each case, the ombud found the department had violated the contract. Although the ombud suspected bad faith, the office didn’t have enough evidence to establish intent.
The ombud recommended PSPC reach an agreement within 60 days to pay damages to the interpreters.
Nearly 200 days later, Cousineau said she and the other interpreters have yet to reach a deal, despite lawyering up.
The problem, she said, is that only the department knows how much money the interpreters would have received had the department followed the rules — and it isn’t willing to share that information.
“The process has been long and painful,” she said.
PSPC spokesperson Mario Morneau said in an email that the department could not provide further details, citing ongoing negotiations.
In the coming months, Cousineau said the cases will move into arbitration-mediation, but she isn’t holding out much hope for a positive result.
Lowest-bid system is the new standard
Christine Pellerin, a freelance interpreter and member of the Canadian section of the International Association of Conference Interpreters, said evidence in the ombud reports “was overwhelmingly damning.”
“I think that the ombud provided a report that went as far as they could because a claim of bad faith is kind of nuclear in government circles,” she said.
Pellerin said the reports dropped shortly after the federal government finalized new procurement rules for freelance interpreters, which make a “lowest-bid system” the new standard.
The association has derided the rules, saying they will result in worse working conditions and lower wages.
“Parliament can’t get enough interpreters, but they’re not creating procurement tools that allow for the industry to sustain itself,” she said."
Ben Andrews
Aug 31, 2026
https://ottawacitizen.com/public-service/interpreters-lost-wages-government
#metaglossia
#metaglossia_mundus
"The future of African languages will not be secured by preserving them only in books, archives and cultural memory. They must also be used in news media, science and public life, and built into the data and artificial intelligence systems shaping the future.
That message took centre stage as the North-West University’s (NWU’s) Indigenous Language Media in Africa (ILMA) research entity marked its 10th anniversary with a two-day symposium at the Birchwood Hotel and OR Tambo Conference Centre in Kempton Park on 25 and 26 August 2026.
Held under the theme “Legends of the past, present and tomorrow”, the symposium explored how indigenous language media can connect historical knowledge with current communication practices and the future of indigenous expression and identity.
ILMA was officially established in 2016. Its research has since covered African-language print, broadcasting and digital media, as well as journalism, democracy, development, political economy and indigenous communication systems.
Language and power
Speaking at the opening, Prof. Dumi Moyo, executive dean of the Faculty of Humanities, said the anniversary celebrated more than ILMA’s first decade. It also recognised a community of scholars whose work has helped establish indigenous language media as a field of academic inquiry.
Prof. Moyo said language remains closely linked to power across Africa, where colonial languages continue to dominate government, education, business and media. Indigenous language media therefore have a role in giving communities greater voice and recognition.
He said ILMA’s work has contributed to decolonising media studies and establishing indigenous languages as legitimate vehicles for knowledge production and public communication. He also acknowledged ILMA director Prof. Abiodun Salawu for helping move indigenous language media from a peripheral concern into a recognised academic field.
Messages of goodwill stream in
Support for ILMA’s work was also reflected during the opening ceremony.
Dr Keaobaka Seshoka, CEO of the Pan South African Language Board, delivered a goodwill message, followed by a goodwill message from Reggy Moalusi, executive director of the South African National Editors’ Forum. Prof. Mirna Nel, deputy dean for research and innovation in the Faculty of Humanities, also delivered a message.
The anniversary’s central challenge came from keynote speaker Prof. Monica Chibita of Uganda Christian University. Her address, “Preservation and more: Indigenous language media, social change and the African Knowledge Society”, argued that indigenous language media should no longer be viewed simply as tools for protecting languages from extinction.
Alive and growing
“Indigenous language media do something more than preserve. They allow people to participate in the present,” she said.
Prof. Chibita said indigenous languages enable people to engage with public policy, health, education, business and knowledge. Their importance will grow as algorithms and AI increasingly influence how people communicate and access information.
She called for “dynamic preservation”, keeping languages alive by using them in investigative journalism, climate science, public health, entrepreneurship, coding, data journalism and artificial intelligence.
Looking towards ILMA’s second decade, Prof. Chibita said African languages must be represented not only in content, but also “in the labs, in the code, in the data, in the standards and in the governance.”
After a decade of building scholarship around indigenous language media, ILMA now faces a new frontier. Its challenge is no longer simply to demonstrate that African languages have a place in modern communication, but to ensure they help shape the technologies, knowledge systems and media platforms of the future."
https://news.nwu.ac.za/ilma-marks-10-years-taking-african-languages-ai-age
#metaglossia
#metaglossia_mundus
Et si demain vous pouviez dialoguer avec une intelligence artificielle en wolof ou en pulaar ? C’est le pari de deux jeunes entrepreneurs sénégalais qui ont développé Awa, un assistant conversationnel capable de comprendre et de répondre dans ces langues africaines. Une innovation qui ambitionne de rendre les outils d’intelligence artificielle plus accessibles aux locuteurs des langues locales. Habib Diao nous fait visiter leur société Andakia avant le lancement officiel de l’outil.
À première vue, elle ressemble à n’importe quelle intelligence artificielle conversationnelle. Mais contrairement à la plupart des outils numériques disponibles aujourd’hui, Awa comprend et répond en Wolof et en Pulaar. Une innovation développée par deux jeunes entrepreneurs sénégalais : Alioune Badara Mbengue et Pape Gorgui Ndoye.
« On a une intelligence artificielle qui s'appelle Awa. Awa se veut être le compagnon, l'ami de tout le monde et qui vous aide à faire tout ce que vous avez besoin de faire. Et cette application là se veut vraiment être ce qu'on appelle une super app. Tout ce dont l'usager a besoin, il doit pouvoir le demander à Awa. Que ce soit une information, que ce soit l'accès à un service, commander à manger, je veux me déplacer, j'ai besoin d'envoyer un colis parce que pour nous, l'intelligence artificielle va devenir une couche, j'ai envie de dire invisible, dans toutes les interactions que les humains ont », présente Alioune Badara Mbengue.
L’idée de création de cette intelligence artificielle est née d’un constat simple, explique-t-il : « AI for development, qui a publié cette étude, montre que la somme de toutes les langues africaines, les données qu'elles représentent sur Internet, c'est moins de 0,02 % de l'ensemble des données disponibles sur Internet. Aujourd'hui, quand vous avez des modèles d'intelligence artificielle, sans en citer, ils s'entraînent justement sur ces données. Et du coup, c'est normal que ces modèles ne soient pas très aptes à s'exprimer dans les langues africaines, justement parce qu'elles sont sous-représentées en termes de données. Donc, c'est le défi principal. Et lorsque nous, on travaille sur ces modèles là, justement, le premier défi est d'avoir de la donnée de qualité, de traiter ces données pour pouvoir entraîner des modèles qui soient aptes à comprendre ces langues-là ».
Avant même son lancement officiel, cette IA suscite déjà un vif intérêt auprès du public. C'est ce qu'affirme Pape Gorgui Ndoye, co-fondateur d’Awa : « Le 18 juillet. On compte enfin lancer Awa. Là on est très optimistes parce qu'on a vu en tout cas beaucoup de retours de beaucoup de gens qui attendent impatiemment l'arrivée d'Awa, qui pourra en tout cas avoir un impact très fort sur leur quotidien ».
Après le Wolof et le Pulaar, les créateurs d’Awa comptent élargir les capacités de leur outil et intégrer d’autres langues africaines. Une étape importante pour cette jeune pousse sénégalaise qui ambitionne de rendre l’intelligence artificielle plus accessible aux locuteurs des langues locales africaines."
Publié le : 01/09/2026 - 00:13
https://www.rfi.fr/fr/podcasts
#metaglossia
#metaglossia_mundus
"When Language Excludes: Linguistic Ostracism, Psychological Ownership, and Destructive Voice Behavior
How language-based exclusion shapes employee behavior and highlights the important role of involvement-oriented work practices in multilingual workplaces
ABSTRACT
Workforces are becoming increasingly multilingual, yet relatively little is known about how language-based exclusion shapes employee behavior. Drawing on social exchange theory, we examine how linguistic ostracism, a subtle form of workplace exclusion, undermines employees' organization-based psychological ownership and subsequently increases destructive voice behavior. We argue that linguistic ostracism weakens employees' felt sense of possession, connection, and responsibility toward the organization, thereby increasing the likelihood of organization-directed retaliatory voice. Furthermore, we propose that perceived high-involvement work practices (HIWPs) buffer these adverse effects by reinforcing employees' sense of value, inclusion, and influence within the organization. Across two studies, we find consistent support for our theorizing. Study 1, a vignette experiment involving 301 employees, demonstrates that linguistic ostracism increases destructive voice behavior. Study 2, a three-wave multisource field study of 241 employee–supervisor dyads, shows that organization-based psychological ownership mediates the relationship between linguistic ostracism and destructive voice behavior. In addition, perceived HIWPs weaken the negative relationship between linguistic ostracism and psychological ownership, thereby reducing the indirect effect of linguistic ostracism on destructive voice behavior. By integrating workplace ostracism, psychological ownership, and perceived HIWPs research, this study advances understanding of how language-based exclusion shapes employee behavior and highlights the important role of involvement-oriented work practices in multilingual workplaces."
Ghulam Murtaza, Rahman Khan, Yasin Rofcanin, Muhammad Zeshan
First published: 30 August 2026
https://doi.org/10.1111/1748-8583.70067Digital Object Identifier (DOI)
EPUB
https://onlinelibrary.wiley.com/doi/full/10.1111/1748-8583.70067
#metaglossia
#metaglossia_mundus
"A Literature Review on Intercultural Competence
As part of the PADMICA project, the consortium, led by ELTE, prepared a comprehensive literature review on interculturality, intercultural competence, and the cultural adaptation of international students. The author of the literature review is Nguyen Luu Lan Anh, a professor at ELTE and a Doctor of the Hungarian Academy of Sciences (MTA).
The literature review provided an important professional background and starting point for the further activities of the project. Among other things, its findings contributed to the development of the PADMICA Intercultural Preparation Module, which aims to support the development of intercultural competence among students participating in international mobility.
The study examines, across four chapters, the factors influencing the cultural adaptation of international students, the role of intercultural competence in this process, and the methods that can be used to support its development.
Main topics covered by the literature review:
International Students and Mobility
The first chapter presents the situation of international students, with particular attention to the importance of preparation before mobility and support following students’ return home.
Cultural Adaptation and Acculturation
The second chapter reviews the literature on acculturation and the cultural adaptation of international students. It addresses, among other things, experiences during mobility and after returning home, as well as factors that may influence the different outcomes of studying abroad.
Intercultural Competence
The third chapter focuses on intercultural competence. It presents its key components and theoretical approaches, and reviews opportunities for its development and assessment, as well as development programmes aimed at international students.
Practical Recommendations for Intercultural Preparation
Drawing on the findings of the literature review, the fourth chapter provides practical recommendations for designing and implementing intercultural preparation programmes.
The literature review therefore not only helped establish a strong professional foundation for the PADMICA project, but also provides consolidated knowledge and practical insights that may be useful to professionals working at higher education institutions in the fields of international mobility and student support."
https://www.elte.hu/en/a-literature-review-on-intercultural-competence
#metaglossia
#metaglossia_mundus
|
Avec la traduction, le dispositif franchit une étape autrement sensible. GoRead décrit une succession d’agents : première version, traitement séparé des noms, lieux et expressions locales, reprise des formulations signalées comme incertaines, révision linguistique, puis arbitrage contextuel.
"Trois mille couronnes suédoises — environ 270 € — pour faire passer un roman de 80.000 mots dans une autre langue. GoRead, la plateforme d’autoédition cofondée par Jonas Tellander, cofondateur de Storytel, ajoute la traduction automatique à sa chaîne de production. Neuf langues sont proposées, sans relecture humaine systématique. Un tarif qui change l’échelle économique de l’opération, sans faire disparaître les questions de droits.
À ce prix-là, GoRead change sérieusement l’échelle annoncée. Selon Boktugg dans son enquête du 7 septembre 2026, le service permet désormais de traduire un manuscrit entre neuf langues : suédois, danois, finnois, islandais, anglais, allemand, polonais, néerlandais et bulgare. Pour un roman d’environ 80.000 mots, la facture tourne autour de 3000 SEK (Couronne Suédoise) par langue, soit quelque 270 €.
Le service de publication de GoRead, lancé en février 2026, aurait déjà fait paraître 250 titres, principalement en Suède, indique Jonas Tellander. La société avait jusque-là concentré son automatisation sur la fabrication du livre : contrôle orthographique, proposition de quatrième de couverture, couverture, EPUB, fichier pour l’impression et narration audio. Sur son propre site, elle revendique 70 % des revenus reversés à l’auteur, qui conserve ses droits sur l’ouvrage.
Avec la traduction, le dispositif franchit une étape autrement sensible. GoRead décrit une succession d’agents : première version, traitement séparé des noms, lieux et expressions locales, reprise des formulations signalées comme incertaines, révision linguistique, puis arbitrage contextuel.
« Cela fonctionne vraiment bien lorsque les différents agents examinent le travail du précédent », assure Tellander. Mais aucune vérification manuelle n’est effectuée par GoRead. L’auteur peut faire relire le résultat avant publication — à ses frais et à son initiative. Le tarif porte donc sur la génération et le traitement automatisé du texte, pas sur une validation humaine.
Neuf langues, et l’audio avec
L’extension linguistique vaut aussi pour le livre audio. GoRead annonce pouvoir produire des narrations synthétiques dans les mêmes neuf langues et distribuer ebooks et audiolivres sur les marchés couverts par Storytel, BookBeat, Nextory et Audible. Au lancement du service, les voix artificielles étaient produites avec la technologie d’ElevenLabs.
C’est du côté d’Audible que l’information mérite une précaution. Boktugg affirme que GoRead est parvenu à faire accepter ses narrations produites par IA, alors que le service n’acceptait jusque-là pas, selon le média suédois, les voix artificielles provenant de producteurs extérieurs. Aucune annonce publique d’Audible consultée ne permet cependant d’en conclure à une ouverture générale de sa plateforme. Le développement doit donc rester attribué à GoRead et à Boktugg.
Spotify devrait suivre, probablement au cours de l’automne. Pour le papier, l’internationalisation demeure plus prosaïque : GoRead continue de travailler avec Publit et ne distribue actuellement les exemplaires imprimés hors de Suède qu’en Finlande et au Danemark, les frais de transport limitant une extension plus large.
Un bouton « traduire » ne cède pas les droits
Reste la question que 3000 couronnes ne règlent pas : a-t-on le droit de faire traduire le texte que l’on vient de déposer ? En Suède, rien n’interdit en soi de recourir à une IA pour traduire un ouvrage. En revanche, l’article 2 de la loi suédoise sur le droit d’auteur réserve au titulaire des droits l’exploitation de l’œuvre, y compris « en traduction ou adaptation ». Un écrivain qui possède encore ses droits de traduction peut donc recourir à un tel outil ; celui qui les a cédés doit d’abord regarder son contrat.
La promesse commerciale de GoRead — l’auteur conserve ses droits face au service — ne répond pas à cette question préalable. Elle décrit la relation avec GoRead, pas les engagements éventuellement conclus avec un éditeur, un agent ou un autre partenaire.
Un second problème apparaît une fois le texte traduit. L’article 4 de la même loi reconnaît des droits à celui qui traduit ou adapte une œuvre, sans lui permettre de porter atteinte à ceux attachés à l’original. Mais le PRV, l’office suédois de la propriété intellectuelle, rappelle que la protection suppose un créateur humain capable d’effectuer des choix libres et créatifs. La Cour de justice de l’Union européenne applique elle aussi ce critère d’originalité.
Pour une version générée intégralement par une machine, sans intervention créative humaine suffisante, l’existence d’une protection autonome demeure donc incertaine. À l’inverse, une reprise substantielle par une personne peut introduire des choix susceptibles d’être protégés. Aucun texte ni aucune décision identifiés ne tranchent spécifiquement le cas des traductions produites par GoRead.
L’AI Act européen ne rend pas davantage le service illégal. Son article 50 prévoit notamment, pour les fournisseurs de systèmes générant du texte ou de l’audio synthétique, des obligations techniques de marquage des sorties, assorties d’exceptions. Mais l’obligation faite aux utilisateurs de signaler un texte produit ou manipulé par IA vise les contenus publiés afin d’informer le public sur des questions d’intérêt public : elle ne crée pas, à elle seule, une mention obligatoire « traduit par IA » sur chaque roman.
GoRead ne précise pas publiquement quel modèle assure ses traductions. Impossible, dans ces conditions, d’attribuer plus précisément les obligations entre les différents acteurs techniques.
Pour les auteurs autoédités — et, demain, les petites maisons auxquelles GoRead veut également ouvrir ses outils — la barrière économique baisse brutalement. La responsabilité éditoriale et juridique, elle, reste entre des mains très humaines.
Nicolas Gary
09/09/2026
Contact : ng@actualitte.com
https://actualitte.com/article/133775/economie/pour-3000-couronnes-goread-traduit-un-roman-par-ia
#metaglossia
#metaglossia_mundus