Toute l'actualité sur http://www.bfmtv.com/ Les spécialistes du dictionnaire Robert et de la grammaire ont validé sa méthode. Elle est déclinée dans plusieur...
Researchers across Africa, Asia and the Middle East are building their own language models designed for local tongues, cultural nuance and digital independence
"In a high-stakes artificial intelligence race between the United States and China, an equally transformative movement is taking shape elsewhere. From Cape Town to Bangalore, from Cairo to Riyadh, researchers, engineers and public institutions are building homegrown AI systems, models that speak not just in local languages, but with regional insight and cultural depth.
The dominant narrative in AI, particularly since the early 2020s, has focused on a handful of US-based companies like OpenAI with GPT, Google with Gemini, Meta’s LLaMa, Anthropic’s Claude. They vie to build ever larger and more capable models. Earlier in 2025, China’s DeepSeek, a Hangzhou-based startup, added a new twist by releasing large language models (LLMs) that rival their American counterparts, with a smaller computational demand. But increasingly, researchers across the Global South are challenging the notion that technological leadership in AI is the exclusive domain of these two superpowers.
Instead, scientists and institutions in countries like India, South Africa, Egypt and Saudi Arabia are rethinking the very premise of generative AI. Their focus is not on scaling up, but on scaling right, building models that work for local users, in their languages, and within their social and economic realities.
“How do we make sure that the entire planet benefits from AI?” asks Benjamin Rosman, a professor at the University of the Witwatersrand and a lead developer of InkubaLM, a generative model trained on five African languages. “I want more and more voices to be in the conversation”.
Beyond English, beyond Silicon Valley
Large language models work by training on massive troves of online text. While the latest versions of GPT, Gemini or LLaMa boast multilingual capabilities, the overwhelming presence of English-language material and Western cultural contexts in these datasets skews their outputs. For speakers of Hindi, Arabic, Swahili, Xhosa and countless other languages, that means AI systems may not only stumble over grammar and syntax, they can also miss the point entirely.
“In Indian languages, large models trained on English data just don’t perform well,” says Janki Nawale, a linguist at AI4Bharat, a lab at the Indian Institute of Technology Madras. “There are cultural nuances, dialectal variations, and even non-standard scripts that make translation and understanding difficult.” Nawale’s team builds supervised datasets and evaluation benchmarks for what specialists call “low resource” languages, those that lack robust digital corpora for machine learning.
It’s not just a question of grammar or vocabulary. “The meaning often lies in the implication,” says Vukosi Marivate, a professor of computer science at the University of Pretoria, in South Africa. “In isiXhosa, the words are one thing but what’s being implied is what really matters.” Marivate co-leads Masakhane NLP, a pan-African collective of AI researchers that recently developed AFROBENCH, a rigorous benchmark for evaluating how well large language models perform on 64 African languages across 15 tasks. The results, published in a preprint in March, revealed major gaps in performance between English and nearly all African languages, especially with open-source models.
Similar concerns arise in the Arabic-speaking world. “If English dominates the training process, the answers will be filtered through a Western lens rather than an Arab one,” says Mekki Habib, a robotics professor at the American University in Cairo. A 2024 preprint from the Tunisian AI firm Clusterlab finds that many multilingual models fail to capture Arabic’s syntactic complexity or cultural frames of reference, particularly in dialect-rich contexts.
Governments step in
For many countries in the Global South, the stakes are geopolitical as well as linguistic. Dependence on Western or Chinese AI infrastructure could mean diminished sovereignty over information, technology, and even national narratives. In response, governments are pouring resources into creating their own models.
Saudi Arabia’s national AI authority, SDAIA, has built ‘ALLaM,’ an Arabic-first model based on Meta’s LLaMa-2, enriched with more than 540 billion Arabic tokens. The United Arab Emirates has backed several initiatives, including ‘Jais,’ an open-source Arabic-English model built by MBZUAI in collaboration with US chipmaker Cerebras Systems and the Abu Dhabi firm Inception. Another UAE-backed project, Noor, focuses on educational and Islamic applications.
In Qatar, researchers at Hamad Bin Khalifa University, and the Qatar Computing Research Institute, have developed the Fanar platform and its LLMs Fanar Star and Fanar Prime. Trained on a trillion tokens of Arabic, English, and code, Fanar’s tokenization approach is specifically engineered to reflect Arabic’s rich morphology and syntax.
India has emerged as a major hub for AI localization. In 2024, the government launched BharatGen, a public-private initiative funded with 235 crore (€26 million) initiative aimed at building foundation models attuned to India’s vast linguistic and cultural diversity. The project is led by the Indian Institute of Technology in Bombay and also involves its sister organizations in Hyderabad, Mandi, Kanpur, Indore, and Madras. The programme’s first product, e-vikrAI, can generate product descriptions and pricing suggestions from images in various Indic languages. Startups like Ola-backed Krutrim and CoRover’s BharatGPT have jumped in, while Google’s Indian lab unveiled MuRIL, a language model trained exclusively on Indian languages. The Indian governments’ AI Mission has received more than180 proposals from local researchers and startups to build national-scale AI infrastructure and large language models, and the Bengaluru-based company, AI Sarvam, has been selected to build India’s first ‘sovereign’ LLM, expected to be fluent in various Indian languages.
In Africa, much of the energy comes from the ground up. Masakhane NLP and Deep Learning Indaba, a pan-African academic movement, have created a decentralized research culture across the continent. One notable offshoot, Johannesburg-based Lelapa AI, launched InkubaLM in September 2024. It’s a ‘small language model’ (SLM) focused on five African languages with broad reach: Swahili, Hausa, Yoruba, isiZulu and isiXhosa.
“With only 0.4 billion parameters, it performs comparably to much larger models,” says Rosman. The model’s compact size and efficiency are designed to meet Africa’s infrastructure constraints while serving real-world applications. Another African model is UlizaLlama, a 7-billion parameter model developed by the Kenyan foundation Jacaranda Health, to support new and expectant mothers with AI-driven support in Swahili, Hausa, Yoruba, Xhosa, and Zulu.
India’s research scene is similarly vibrant. The AI4Bharat laboratory at IIT Madras has just released IndicTrans2, that supports translation across all 22 scheduled Indian languages. Sarvam AI, another startup, released its first LLM last year to support 10 major Indian languages. And KissanAI, co-founded by Pratik Desai, develops generative AI tools to deliver agricultural advice to farmers in their native languages.
The data dilemma
Yet building LLMs for underrepresented languages poses enormous challenges. Chief among them is data scarcity. “Even Hindi datasets are tiny compared to English,” says Tapas Kumar Mishra, a professor at the National Institute of Technology, Rourkela in eastern India. “So, training models from scratch is unlikely to match English-based models in performance.”
Rosman agrees. “The big-data paradigm doesn’t work for African languages. We simply don’t have the volume.” His team is pioneering alternative approaches like the Esethu Framework, a protocol for ethically collecting speech datasets from native speakers and redistributing revenue back to further development of AI tools for under-resourced languages. The project’s pilot used read speech from isiXhosa speakers, complete with metadata, to build voice-based applications.
In Arab nations, similar work is underway. Clusterlab’s 101 Billion Arabic Words Dataset is the largest of its kind, meticulously extracted and cleaned from the web to support Arabic-first model training.
The cost of staying local
But for all the innovation, practical obstacles remain. “The return on investment is low,” says KissanAI’s Desai. “The market for regional language models is big, but those with purchasing power still work in English.” And while Western tech companies attract the best minds globally, including many Indian and African scientists, researchers at home often face limited funding, patchy computing infrastructure, and unclear legal frameworks around data and privacy.
“There’s still a lack of sustainable funding, a shortage of specialists, and insufficient integration with educational or public systems,” warns Habib, the Cairo-based professor. “All of this has to change.”
A different vision for AI
Despite the hurdles, what’s emerging is a distinct vision for AI in the Global South – one that favours practical impact over prestige, and community ownership over corporate secrecy.
“There’s more emphasis here on solving real problems for real people,” says Nawale of AI4Bharat. Rather than chasing benchmark scores, researchers are aiming for relevance: tools for farmers, students, and small business owners.
And openness matters. “Some companies claim to be open-source, but they only release the model weights, not the data,” Marivate says. “With InkubaLM, we release both. We want others to build on what we’ve done, to do it better.”
In a global contest often measured in teraflops and tokens, these efforts may seem modest. But for the billions who speak the world’s less-resourced languages, they represent a future in which AI doesn’t just speak to them, but with them."
"UNM professor pushing forward research into sign language
ALBUQUERQUE, N.M. (KRQE) – A linguistics professor at the University of New Mexico is helping shape the future of sign language research, using unique teaching methods that help her students re-imagine communication.
Professor Erin Wilkinson is one of only a handful of deaf tenure-track professors in the United States with a focus on bilingualism. Her courses challenge assumptions about language and communication and how learning about these topics isn’t limited to spoken language. “So, there’s a lot to be learned from Deaf folks, how people learn the cues that make for effective communication, even though maybe other people don’t think it’s effective, but they’re obviously doing something right. I do think as a university, and also even outside of the university, we can learn so much about how we can connect,” explained Professor Erin Wilkinson.
As a deaf Professor Wilkinson, teaching hearing students, she believes it helps them better understand other communication strategies. She says that deaf people early on have to develop strong communication skills — growing up in a world where everyone is assumed to be hearing.
That’s why Wilkinson offers some of her courses, like ‘Structure of ASL’, without interpreters. “They would come into the classroom, and everyone would expect their teacher to be hearing and to speak English. And then they sit down, and they go, What, who are you? You’re not talking, you’re signing. And I think that’s such an important experience for students to have, to increase their understanding of the world and communication as a whole,” emphasized Wilkinson.
She said that doing this mirrors the challenges that deaf individuals face, especially in academic settings. Wilkinson believes that this model she uses in her classroom can offer greater insight into communication..." by: Melissa Torres Posted: Dec 25, 2025 / 09:58 PM MST Updated: Dec 26, 2025 / 10:12 AM MST https://www.krqe.com/news/albuquerque-metro/unm-professor-pushing-forward-research-into-sign-language/ #Metaglossia #metaglossia_mundus
"From phishing to bias: Study maps the hidden threats behind large language models
FAYETTEVILLE, GA, UNITED STATES, December 25, 2025 /EINPresswire.com/ -- Large language models (LLMs) have become central tools in writing, coding, and problem-solving, yet their rapidly expanding use raises new ethical and security concerns. The study systematically reviewed 73 papers and found that LLMs possess dual roles—empowering innovation while simultaneously enabling risks such as phishing, malicious code generation, privacy breaches, and misinformation spread. Defense strategies including adversarial training, input preprocessing, and watermark-based detection are developing but remain insufficient against evolving attack techniques. The research highlights that the future of LLMs will rely on coordinated security design, ethical oversight, and technical safeguards to ensure responsible development and deployment.
Large language models (LLMs) such as generative pre-trained transformer (GPT), bidirectional encoder representations from transformers (BERT), and T5 have transformed sectors ranging from education and healthcare to digital governance. Their ability to generate fluent, human-like text enables automation and accelerates information workflows. However, this same capability increases exposure to cyber-attacks, model manipulation, misinformation, and biased outputs that can mislead users or amplify social inequalities. Academic researchers warn that without systematic regulation and defense mechanisms, LLM misuse may threaten data security, public trust, and social stability. Based on these challenges, further research is required to improve model governance, strengthen defenses, and mitigate ethical risks.
A research team from Shanghai Jiao Tong University and East China Normal University published (DOI: 10.1007/s42524-025-4082-6) a comprehensive review in Frontiers of Engineering Management, (2025) examining ethical security risks in large language models. The study screened over 10,000 documents and distilled 73 key works to summarize threats such as phishing attacks, malicious code generation, data leakage, hallucination, social bias, and jailbreaking. The review further evaluates defense tools including adversarial training, input preprocessing, watermarking, and model alignment strategies.
The review categorizes LLM-related security threats into two major domains: misuse-based risks and malicious attacks targeting models. Misuse includes phishing emails crafted with near-native fluency, automated malware scripting, identity spoofing, and large-scale false information production. Malicious attacks appear at both data/model level—such as model inversion, poisoning, extraction—and user interaction level including prompt injection and jailbreak techniques. These attacks may access private training data, bypass safety filters, or induce harmful content output.
On defense strategy, the study summarizes three mainstream technical routes: parameter processing, which removes redundant parameters to reduce attack exposure; input preprocessing, which paraphrases prompts or detects adversarial triggers without retraining; and adversarial training, including red-teaming frameworks that simulate attacks for robustness improvement. The review also introduces detection technologies like semantic watermarking and CheckGPT, which can identify model-generated text with up to 98–99% accuracy. Despite progress, defenses often lag behind evolving attacks, indicating urgent need for scalable, low-cost, multilingual-adaptive solutions.
The authors emphasize that technical safeguards must coexist with ethical governance. They argue that hallucination, bias, privacy leakage, and misinformation are social-level risks, not merely engineering problems. To ensure trust in LLM-based systems, future models should integrate transparency, verifiable content traceability, and cross-disciplinary oversight. Ethical review frameworks, dataset audit mechanisms, and public awareness education will become essential in preventing misuse and protecting vulnerable groups.
The study suggests that secure and ethical development of LLMs will shape how societies adopt AI. Robust defense systems may protect financial systems from phishing, reduce medical misinformation, and maintain scientific integrity. Meanwhile, watermark-based traceability and red-teaming may become industry standards for model deployment. The researchers encourage future work toward AI responsible governance, unified regulation frameworks, safer training datasets, and model transparency reporting. If well-managed, LLMs can evolve into reliable tools supporting education, digital healthcare, and innovation ecosystems while minimizing risks linked to cybercrime and social misinformation.
Human speech leaves no fossils. The origin of languages remains one of science’s greatest mysteries, explored by modern linguistics.
"How Did Language Begin? The Great Mystery of Human Communication
This is a great human strength – we can encode the past, we can imagine the future and talk about it. And this happened precisely because we use language. The productivity of language detaches us from the "here and now," explains Prof. Aneta Wysocka from the Maria Curie-Skłodowska University in Lublin in an interview for Holistic News.
The origin of languages: One of the eternal mysteries
Maria Mazurek: How did languages originate?
Prof. Aneta Wysocka:* This question intrigued the greatest minds long before linguistics even existed.
What is the answer?
We don’t know it. Speech leaves no fossils. The only evidence of the shape of ancient languages is writing. But a vast amount of time passed between the moment natural languages were born and the moment people began to record them. We are forced to rely on guesses—or perhaps on tracking traces.
What kind of traces?
We have certain anthropological clues: the position of the larynx or the size of the brain estimated based on the shape and size of the skull. Based on this, we can formulate hypotheses that humans became capable of very precise articulation hundreds of thousands of years ago. And that was necessary to use language as we understand it today.
What is language? Describing reality through symbols
I asked about the origin of languages. What exactly is language?
I will answer from the perspective of a linguist – a productive system of symbolic signs.
Could you tell us how to understand that?
To put it simply: from a limited repertoire of symbols—or, simplified, words that we use to name things and phenomena—we are able to generate an unlimited number of sentences describing various states of affairs. Those that occurred in the past, those that exist now, those that might exist in the future, and those we can only imagine.
Is this possible thanks to grammar?
Yes. Grammar is nothing more than a rule for combining symbols into utterances. It enables this “productivity.” There are many communication systems—both those that evolved naturally and those created by humans—but only natural language meets the condition of productivity. We know that animals have ways of communicating (more or less advanced, more or less understood), but we have no evidence that they meet the condition of productivity.
The origin of languages: Protolanguage vs. instinct
So we don’t know how the origin of languages occurred. Do we know if they all come from a common “protolanguage,” or did they develop independently of each other?
We have alternative hypotheses. And too little knowledge to grant priority to one of them.
If it’s true that languages were born independently, could they be “hardwired” into our brains?
There is such a hypothesis popularized by cognitive scientist Steven Pinker. Pinker wrote a book titled The Language Instinct, which contains the thesis (well-documented, by the way) that using language is an instinct in humans—an innate and not necessarily conscious mechanism.
So he places it alongside maternal instinct, self-preservation, or reproductive instinct?
Yes. Pinker believes that language is biologically “built-in” to us.
And what do you think?
As a scientist, I try not to favor any hypotheses as long as data is lacking. And in this case, we still know too little.
In a child, on paper, on screen, and through signs
What do we know then?
We certainly know that young children have a natural inclination and need to acquire the natural language spoken by those around them. They want and are able to learn incredibly quickly which categories are worth dividing the world into to navigate it better. To name what surrounds them. They also have a need to communicate with another human being. We also know that there is a developmental window of a few years during which a small person should “fit in” to acquire language. Later, it becomes much more difficult, though not impossible.
Does language have to be verbal?
Verbum means word. A word is a symbol—a symbol we have in our minds. It is a secondary matter how it manifests itself. It can manifest as speech—a stream of air shaped by our articulatory apparatus—or it can manifest as a sequence of signs on writing material—on a computer screen or a piece of paper written with a pen. But it can also manifest as a symbolic gesture. We deal with symbolic gestures in the case of sign language. There is no doubt that it is a productive language—one can express all our thoughts about what was, is, and will be through it.
The origin of languages and pre-linguistic reality: Lost paradise or utopia?
What was, is, and will be – is that the key?
Yes. This is a great human strength – we can encode the past, we can imagine the future and talk about it. And this happened precisely because we use language. The productivity of language detaches us from the here and now.
You called it a human strength. Others would call it a curse. After all, mindfulness, meditation, yoga, certain forms of therapy – all of these are meant to help us be “here and now.” To stop, if only for a moment, this chase of thoughts – into the future, the past, through various threads and topics.
Hence the utopian idea of returning to nature. We have a suspicion—or perhaps we would very much like it to be so—that returning to primal forms of existence would ensure our well-being. But would it really be so? I don’t know, but I don’t think I would give up the ability to look into the future and explore the past just to feel good.
How do we think? Words vs. images
Me neither. But I would like to stop the flood of thoughts sometimes.
Like most people.
And are thoughts always language? Are there non-verbal thoughts?
Of course, there are. Cognitive psychology distinguishes between different types of thinking. Engineers, when solving spatial problems, do not have to use language in that process. Not all thinking is linguistic. However, language, words, and linguistic symbols are anchors for thoughts.
I asked this question because it seems to me that my thoughts are always words. But when I asked others about it, the answers were different.
A great asset of our species is neurodiversity. Because we are different, yet function within a community, we can use these specific traits for the benefit of all. There are engineers who think spatially, who can mentally rotate geometric figures, which I can’t even imagine. And there are those who think mainly through language, and thanks to them, we have wonderful literature.
The origin of languages and the interpretation of reality
And does the language we think in determine our way of understanding the world?
I wonder if “determines” is the best word. I think I would prefer to say: facilitates.
Why?
Let me explain: language contains certain interpretations of reality made by our ancestors that proved useful. So we, by learning words, also learn models of interpreting the world. The first language—which will always remain the most important one—organizes our cognitive system. Of course, every language is productive, so we can express everything in it. But the language we use (especially the first one) directs our attention to those categorizations that are already ready, served to us on a platter. It suggests an interpretation of the world. We can be satisfied with that, or we can look for alternative visions of reality—and still use the same language. So it is always a matter of cognitive activity. That’s why I wouldn’t speak of determinism here, but rather of what is easier and what is more difficult. And whether a person wants to take the trouble to see the world differently.
There are elementary words found in every language in the world. However, most words are specific—specific to a given language and culture. I recall a beautiful lecture by Prof. Anna Wierzbicka about the word pamiątka (souvenir/keepsake), which the professor considers a word specific to the Polish language. Although we can translate it as “souvenir,” we lose what it carries with it. The word pamiątka shows our culture’s established attitude toward the past, toward time. Even toward other people.
New languages reveal other images of the world
There are also words that are completely untranslatable.
Perhaps I should put it differently: translatable, but having no single-word equivalents in other languages. The German Reisefieber can be translated as “a state of nervousness before a journey,” but we need a whole phrase for that. Finns have one word for the heat given off by stones in a sauna. These are anecdotal examples—interesting, sometimes funny—but they show that our culturally conditioned interpretations of the world differ. So learning a second language is not only about acquiring tools for communication but also new tools for thinking. A new language opens up a new image of the world for us.
* Dr. hab. Aneta Wysocka, Professor at the Maria Curie-Skłodowska University in Lublin – linguist, academic teacher, and member of the Program Board and Board of Directors of the “Akcent” Eastern Cultural Foundation. She is the co-founder and scientific editor of the linguistic yearbook “Idiolekty” (Idiolects). Her honors include the Medal of the President of the City of Lublin and the Cross of Merit of the Republic of Poland. She is the author of numerous books and scientific publications on language.
Read the original article in Polish: Wciąż nie wiemy, jak powstają języki. Prof. Wysocka szuka odpowiedzi
"HUAIN’s Cutting-Edge Interpreter Systems Set New Standards for Global Conference Communication News Provided By aicc December 24, 2025, 13:35 GMT
HUAIN's Cutting-Edge Interpreter Systems
China Top Wired Microphone Factory - HUAYIN
ZHUHAI, GUANGDONG, CHINA, December 24, 2025 /EINPresswire.com/ -- In today’s interconnected business environment, international collaboration has become the foundation of progress. Whether in diplomacy, global trade, or corporate summits, communication across multiple languages determines the success of every meeting. Yet, language barriers remain a major challenge. That’s why HUAIN’s Cutting-Edge Interpreter Systems from HUAIN are redefining how multilingual communication takes place at global events — enabling seamless understanding and real-time translation in the most demanding environments.
A New Era of Global Conference Communication As international cooperation grows, so does the demand for advanced interpretation technology. Traditional systems are no longer sufficient for today’s complex events that combine in-person and remote participation. From government meetings to academic forums and corporate expos, organizers need technology that ensures clarity, accuracy, and reliability across languages and distances. In this context, HUAIN has emerged as a Global Leading AV Equipment Supplier—offering state-of-the-art interpretation solutions built on innovation, scalability, and superior sound performance. Its products have set a benchmark for how modern conferences manage multilingual interaction, combining advanced digital processing with user-friendly design and robust infrastructure.
What Defines a Next-Generation Interpreter System? A truly world-class interpreter system must meet several demanding criteria. Beyond basic audio clarity, it must provide stable signal transmission, support for multiple channels, and secure encryption to ensure confidentiality. At international events, interpreters need instantaneous, interference-free communication that doesn’t falter under heavy data traffic. Modern interpretation systems should also integrate with a wider conference ecosystem, including wired and wireless microphones, voting units, digital discussion systems, and remote conferencing platforms. This is where HUAIN’s solutions excel. They blend cutting-edge digital infrared and wireless communication technologies with intelligent control interfaces, offering a holistic platform rather than just a stand-alone device. Furthermore, buyers today seek flexibility — systems that can adapt to diverse event scales, from small bilingual meetings to large summits with 16 or more simultaneous languages. HUAIN’s interpreter systems deliver this adaptability, making them ideal for a globalized communication landscape.
HUAIN: The Innovator Behind Intelligent Conference Communication Founded in 2015, HUAIN is a national high-tech enterprise that has quickly become one of the China Top Wired Microphone Factory leaders in the audio-visual conferencing industry. The company focuses on developing integrated conference systems, including intelligent digital discussion units, paperless meeting systems, visual management platforms, and interpreter systems. What sets HUAIN apart is its continuous investment in R&D and its pursuit of practical innovation. Its proprietary technologies—such as dual-diaphragm sound pickup and anti-recording ultrasonic diffraction—have revolutionized how audio is captured and transmitted in professional settings. This engineering excellence ensures precise voice reproduction and crystal-clear interpretation, even in complex acoustic environments. With more than 10,000 successful installations across government institutions, educational organizations, and multinational corporations, HUAIN has established itself as one of the China Best AV Solution providers for global communication.
Inside HUAIN’s Interpreter Systems: Technology That Speaks Every Language HUAIN’s interpreter systems are designed to meet international standards while maintaining exceptional usability and reliability. They feature:
Multi-channel translation capabilities: Up to 16 simultaneous languages supported, ensuring real-time comprehension across diverse audiences. Digital IR and RF technology: Infrared systems eliminate interference, providing privacy and signal stability in large-scale events. Crystal-clear audio performance: 48kHz digital sampling rate and >90dB signal-to-noise ratio ensure natural voice quality. Smart interpreter console design: Ergonomic controls allow interpreters to switch languages, mute microphones, and adjust sound levels with precision. Scalable system architecture: From small seminars to international summits, HUAIN’s systems can expand without performance loss. Seamless integration: Fully compatible with HUAIN’s conference microphones, paperless meeting systems, and visual management tools. These features transform conference interpretation from a functional necessity into an effortless experience. They help eliminate the delays and distortions common in older systems while maintaining the integrity and privacy of communication.
Practical Applications and Global Recognition HUAIN’s interpreter systems have been deployed in diverse scenarios: international forums, diplomatic summits, educational symposiums, and multinational corporate conferences. For example, several major government offices and universities in Asia and Europe have adopted HUAIN’s solutions to facilitate multilingual collaboration and policy discussions. Such real-world deployments underline the systems’ reliability, scalability, and superior sound engineering. Clients appreciate that HUAIN not only supplies products but also provides end-to-end AV integration—installation, configuration, training, and maintenance. This comprehensive approach has positioned the company as a trusted Global Leading AV Equipment Supplier capable of supporting mission-critical communication worldwide.
The Competitive Edge: Why HUAIN Stands Out HUAIN’s strengths extend far beyond product specifications. The company has built its success on three core pillars:
Integrated System Design – By offering wired and wireless microphones, digital discussion systems, paperless meeting solutions, and interpreter systems, HUAIN delivers one cohesive ecosystem that ensures compatibility and ease of operation. Global Quality, Local Manufacturing – Combining Chinese production efficiency with international quality control, HUAIN delivers advanced systems that meet global certification standards such as CE and FCC. Customer-Centric Service – HUAIN provides localized after-sales support, quick response times, and flexible customization options tailored to clients’ needs.
This customer-first philosophy reflects HUAIN’s long-term commitment to reliability and partnership—key qualities that buyers seek when choosing suppliers for complex conference infrastructure.
Industry Trends: The Future of Interpretation Technology As hybrid meetings and digital events become mainstream, the need for intelligent interpreter systems is expanding rapidly. The integration of AI-assisted translation, network-based management, and smart analytics will soon reshape the industry. HUAIN is already at the forefront of this transformation, integrating digital processing and visual management with its core conference systems. Moreover, with globalization accelerating, multilingual communication will remain indispensable. Enterprises, international organizations, and educational institutions are increasingly turning to suppliers capable of delivering comprehensive AV ecosystems rather than isolated devices. HUAIN’s technological maturity and broad product portfolio place it in a prime position to meet these evolving expectations.
Building Bridges Through Innovation Effective communication lies at the heart of international collaboration, and HUAIN’s interpreter systems are making that communication smarter, faster, and clearer than ever before. By combining precision engineering, scalable system design, and a deep understanding of real-world conferencing needs, HUAIN (https://www.huainpro.com/) is not only setting new benchmarks for the industry but also shaping the future of global meeting technology. For organizations seeking reliability, clarity, and innovation in multilingual communication, HUAIN’s cutting-edge interpreter systems offer the ideal balance between performance and value — a solution that truly sets new standards for global conference communication. HUAIN HUAIN +86 756 863 2108 info@sustainwastesolution.com
"Et si la traduction devenait enfin un métier aussi contrôlé qu’une opération chirurgicale Dans la justice, la santé ou la recherche, une simple phrase mal traduite peut tout faire basculer.
Et si la traduction devenait enfin un métier aussi contrôlé qu’une opération chirurgicale ou un audit financier ?
C’est précisément ce qu’a accompli AFTCom. L’agence française vient d’obtenir la certification ISO 17100, la référence mondiale qui audite désormais le métier de traducteur avec la même rigueur que les industries réglementées.
Concrètement, chaque étape est passée au crible : qualification des traducteurs, double révision systématique, traçabilité de chaque décision, supervision du moindre mot livré. La traduction devient un processus certifié, plus proche de la science que de l’artisanat.
Ce bouleversement, peu visible du grand public, redéfinit une profession souvent sous-estimée. Dans les hôpitaux, les tribunaux ou les laboratoires, ces mots validés et vérifiés deviennent des garanties de fiabilité, parfois de sécurité.
Derrière cette démarche se trouve un homme qui a lui-même redéfini la notion de rigueur. Malvoyant, diplômé de l’ESIT, Mohammed Bourasse a bâti en dix ans une agence inclusive de 90 collaborateurs, en faisant de la précision linguistique une mission d’utilité publique.
"CAIRO, Dec 25 (KUNA) -- The Arab League, in collaboration with the Arab Organization for Dialogue and International Cooperation, hosted an intellectual conference under the theme: "Image of Arabs and Interculture Dialogue..Future Visions", on Thursday..."
"Spines introduces AI-powered voice cloning feature for audiobook production
Venture capitalist Michael Eisenberg, a backer of Spines through Aleph, tested its AI voice cloning on his upcoming book - his wife couldn’t tell it wasn’t him
ynet Global | published:12.24.25
AI-powered self-publishing platform Spines introduces author voice cloning for audiobooks, expands translation services to seven languages, and grows its author base to more than 6,000 users.
The company launched in 2021 as a spinoff of Niv Publishing, Israel's main self-publishing platform. Using AI, the platform automates multiple aspects of publishing, including editing, proofreading, cover design, marketing, and distribution across more than 100 retailers. Authors upload manuscripts and receive published books within three weeks...
In addition to book production, authors can now create full-length audiobooks narrated in their own voice by providing 33 minutes of audio samples, which is a significant reduction from the 40-plus hours traditionally required for studio recording. The voice cloning feature builds on the company's existing audiobook service, which now offers more than 300 artificial voice options. The platform has also implemented AI-assisted proofing.
"We can create a voice model and do the whole audiobook from that. Our tools can predict where a recording may be at fault, directing where someone can go in and make changes manually. We are now able to save a lot of human labor in creating those audiobooks", says Yehuda Niv, Spines CEO and founder...
Venture capitalist Michael Eisenberg, investor at Aleph and one of Spines' backers, tested the technology for his forthcoming book. "He sent it to his friends and he told them, 'What do you think about when you're recording?' They actually didn't know it's an AI," Niv said. "Even his wife couldn't tell the difference."
Beyond audiobooks, Spines has implemented an AI-human hybrid approach to translation services, supporting Spanish, French, German, Italian, Polish, Portuguese, and additional languages. "AI for speed, human for quality, because the translations based on AI are not good. They are not that accurate yet," Niv explained. "So we have a man in the loop who reviews. The translation service represents Spines' effort to help authors double and quadruple their audience potential because the book is in more languages and in more formats," according to Niv.
Among those backing the company is Oren Zeev, founder and managing partner of Zeev Ventures and an early investor in Audible. "Back in 2003, I saw the untapped potential of audiobooks with Audible, long before they became popular," Zeev said. "Today, I see a similar transformative power in Spines to reshape the book publishing industry."
Looking ahead, Spines is developing a copilot tool to help authors complete manuscripts. "We want to help people to not only publish their book, but also finish writing their manuscript, because a lot of authors are getting stuck," Niv said."
"Western Armenian Translation Grant Awarded to Aktokmakyan and Janbazian
This year’s grant covered literary works written in Western Armenian from any period and across all literary genres. Among the submissions received, two works were found equally successful by the judges.
The Israelyan Western Armenian Translation Grant, which aims to ensure more Armenian literature is translated into English and from English into Western Armenian each year, has found its recipients.
This year’s grant covered literary works written in Western Armenian from any period and across all literary genres. Among the submissions received, two works were found equally successful by the judges. In line with IALA’s vision to make Armenian literature available to a broader audience, it was decided to award the grant to two separate works this year.
The International Armenian Literary Alliance (IALA) extended its thanks to all applicants of the 2025 Israelyan Western Armenian Translation Grant and to judges Tamar Marie Boyadjian and Jennifer Manoukian. The grant, which draws a wide range of submissions every year, once again facilitated the evaluation of diverse and ambitious works.
Heranoush Arshagyan’s Lusnyag «Լուսնեակ», translated by Maral Aktokmakyan, and Souren Chekijian’s Half-Drawn «Անաւարտ դիմանկար», translated by Rupen Janbazian, were deemed worthy of the award.
The Winners: Maral Aktokmakyan: Dr. Maral Aktokmakyan is a scholar, writer, and translator whose work centers on Western Armenian literature and its enduring creative legacy. She contributes to this legacy through her original writing, crafting essays, analysis, and creative works that illuminate overlooked voices and literary traditions. She makes Armenian literature accessible to wider audiences through her translations between Armenian, English, and Turkish, preserving the richness of the original texts. Her research appears in both academic and literary venues, where she examines how literature carries memory, imagination, and identity into new contexts. Through her writing, scholarship, and translation efforts, Dr. Aktokmakyan is committed to expanding the visibility, vitality, and future possibilities of Western Armenian literary art.
Rupen Janbazian: Rupen Janbazian is a writer, editor, and translator from Toronto, currently based in Yerevan. He is the editor of Torontohye, a bilingual Armenian-English community newspaper in Toronto, and the former editor of the Armenian Weekly. Writing in both English and Western Armenian, Janbazian’s work often explores questions of homeland-diaspora, identity, and community life between Canada and Armenia. Janbazian has translated, co-translated, and edited a range of literary works, memoirs, articles, and short fiction. He is currently working on an English translation of his late friend Souren Chekijian’s Western Armenian novel «Անաւարտ դիմանկար», set in Toronto, which examines exile, aging, desire, and the inner world of a Lebanese-Armenian painter in Canada. He lives in Yerevan with his partner Araz, and their dog, Srjeni.
Souren A. Israelyan and the Western Armenian Translation Grant The 2025 Israelyan Western Armenian and English Translation Grants from the International Armenian Literary Alliance are made possible by a generous donation from Souren A. Israelyan. This fund aims to support the translation of more Armenian literature from and into English.
Souren A. Israelyan has been practicing law since 2003 and founded his own law office in 2008. His firm serves clients who have been seriously injured, handling matters from initial investigation through court proceedings, including trials and appeals.
"The Bosnian-Persian dictionary, which was created in the organization of the Scientific Research Institute “Ibn Sina” and the Faculty of Philosophy of the University of Sarajevo, was presented on Wednesday in Gazi Husrev-bey’s library.
One of the authors of the dictionary, professor of Persian language and literature at the Department of Oriental Philology of the Faculty of Philosophy of the University of Sarajevo, Munir Drkić, pointed out that creating the dictionary was an extremely demanding job, especially because it was being translated from Bosnian to Persian.
“Such a dictionary is needed by our culture because it connects the Bosnian language with the long lexicographical tradition of the Persian language. This is the first Bosnian-Persian dictionary and one of the capital projects in connecting two lexicographic traditions,” said Drkić.
He added that the agreement on the creation of the dictionary was reached in 2011, and the work began in February 2012, and lasted for almost 14 years.
The dictionary contains 50,000 entries, along with translations, transcriptions, examples, proverbs and phrases, and was published with the support of the Federal Ministry of Education and Science.
“The Persian language, although it comes from a distant geographical area, is typologically similar to the Bosnian language and had a significant influence in the Balkans, especially in the Ottoman period,” emphasized Drkić.
President of the Foundation for Iranian Studies of the Islamic Republic of Iran, dr. Ali Akbar Salehi assessed that the Persian language has a special place in the cultural history of Bosnia and Herzegovina.
“The Persian language can also represent a bright thread connecting two cultures in the future. Bosnia and Herzegovina, as a link between East and West, occupies a special place in Europe, and the Persian language, one of the pillars of Eastern culture, is spread from China to Europe,” said Salehic, Fena writes.
The Bosnian-Persian dictionary is the result of more than 14 years of continuous scientific effort by a group of professors and researchers of the Persian and Bosnian languages. The project was started in 2011 in Sarajevo under the leadership of eminent prof. Ph.D. Namir Karahalilović with the participation of a group of professors and researchers of Persian language and literature.
Previously, in 2010, a Persian-Bosnian dictionary was published by the “Ibn Sina” Scientific Research Institute. The Bosnian-Persian dictionary completes this previous Dictionary and represents an additional step in promoting education and research in these two languages.
In the introduction to the Dictionary, the respected authors presented the scientific and technical structure of this work and analyzed in detail the history of the Persian language and literature in Bosnia and Herzegovina and its role in the cultural interactions of the two nations.
The president of the Academy of Persian Language and Literature IR Iran, dr. Haddad Adel, and the reviewers of the dictionary are prof. Ph.D. Taghi Poornamdarian, prof. Ph.D. Ismail Palić and Prof. Ph.D. Ahmed Zildzic."
Ghana's Komba people received their complete Bible after 16 years of dedicated translation work. Speaking during the launch event last month, Rev. Dr. John Kwesi Addo Jnr., General Secretary of the Bible Society of Ghana (BSG), emphasized that this Bible will deepen faith while serving as a major repository to preserve the Komba language and culture from extinction.
"This was more than a dedication. It was a cultural event that united the entire community: Christians, Paramount Chiefs, elders, and Muslims, all celebrating 16 years of tireless translation work," BSG noted in a statement.
BSG partnered with the Lutheran Bible Translators to deliver the Komba Bible in the hope that the Bible “will shape lives, strengthen families, and combat moral decay.”
The Komba people live in Ghana's northeast region. Unlike many neighboring groups with centralized chieftaincy structures, the Konkomba traditionally organized themselves without a central ruling authority. Social life centers around lineages, clans, village elders, and local religious and spiritual leaders.
Stay informed with The Christian Daily Newsletter
Sign up
Historically, their worldview embraced traditional spiritual beliefs: reverence for ancestral spirits, belief in natural spirits inhabiting rivers, trees, and land, plus rituals led by traditional healers or priests.
Many Konkomba have since adopted Christianity, while others follow Islam. Traditional beliefs and practices still influence some communities.
According to Lutheran Bible Translators, Baptist missionaries established the first congregation among the Komba in Namong during the 1950s. In the early 1980s, Lutheran missionaries Tim and Beth Heiney moved to Ghana to serve the Konkomba area.
In 1968, church authorities assigned Reverend Walter Demoss and his wife Helena to plant churches and train local leaders in Northern Ghana. Though called specifically to serve the Moba people, Reverend Demoss also mentored a young Komba man, Reverend Samuel Konlaan.
Rev. Konlaan later voiced concern that the only existing Bible translation remained difficult for his people to understand because of the many dialects within the Komba language.
📱 Get Breaking News on WhatsApp
Join our WhatsApp channel for instant updates on Christian news worldwide
After years of preparation, organizers fully launched the New Testament translation effort in 2005 and assembled the team to translate the New Testament. Members included Mr. Elijah Matibin, project coordinator with expertise in Scripture engagement and literacy; Mr. David Federwitz, literacy and Scripture engagement advisor; Rev. Samson Bilafanim, translator; Rev. Emmanuel Mananyina, translator; Mr. James Adongo Wajak, translator; Rev. Nathan Esala, linguist and translation advisor; and Dr. Fabian Dapila, translation consultant.
Lutheran Bible Translators reports that the community played a vital role in the translation, ensuring the work not only conveyed God's Word but addressed the community's needs.
"The translation team sends them a printed copy. In some translation projects, reviewers choose to meet in a group, but the Komba reviewers decided to make their suggestions independently," LBT explained.
On November 1, 2014, the Komba community gathered for a joyful celebration when they finally received the New Testament. Lutheran Bible Translators
On Nov. 1, 2014, the Komba community gathered for a joyful celebration when they finally received the New Testament. Rev. Mananyina expressed his delight: "Reading the Bible has become part of my people. They are reading it day in and day out, and they have taken it upon themselves to do so. They have learned to read, and now they can go out and preach because they can read the Bible, something they were unable to do in the past."
Soon after the dedication, work on the Old Testament translation began in 2015. Mr. Elijah Matibin took on leadership of the project as KOLIBITRAP Coordinator. Lutheran Bible Translators, the Evangelical Lutheran Church of Ghana, KOLIBITRAP, and the Bible Society of Ghana signed a Memorandum of Understanding (MOU) to launch the Old Testament translation effort.
During this period, the team also recorded the New Testament in audio form and integrated it with the text to create a smartphone app. One Way Africa has now produced the full Bible in audio format to enhance Bible engagement.
Earlier on Nov. 2, the Bible Society of Ghana launched the Dagaare Bible after 18 years of work. Today, many Dagaare people, located in North West Ghana, are Catholic or other Christian denominations, while a number also practice Islam.
"The vast grounds of St. Andrew's Cathedral overflowed with people from every corner of the Upper West Region: men, women and children, Reverend Ministers, and the Regional Minister, all eager to witness this historic moment," the society reported.
In a 2023 report, BSG revealed that lack of financial support hinders their work to translate the Bible into numerous languages. The organization estimates that translating a single verse costs $20, bringing the overall cost to complete one language to $622,040 (GH¢7.2 million) over 10 to 15 years.
A new MA Interpreting course at Queen's University is preparing students for the increasing demand for interpreters.
"An experienced sign language interpreter has said people like her can experience "emotional whiplash" trying to manage the demands of the role with limited support.
Dr Sally Gillespie is the project coordinator of a new MA Interpreting course at Queen's University Belfast (QUB) which is preparing students for the increasing demand for interpreters.
There are 44 registered Sign Language Interpreters in Northern Ireland, however, she said this is not reflective of the number of working interpreters.
Dr Gillespie said issues with availability can mean they are often called into highly emotional situations at short notice, sometimes moving between a christening and a life-or-death situation in hospital.
"Demand has always outstripped supply," Dr Gillespie said.
"The opportunity to grow and invest in the next generation of interpreters has always been a struggle."
In February, the Sign Language Bill was introduced at Stormont to officially recognise and promote British Sign Language (BSL) and Irish Sign Language (ISL) in Northern Ireland.
Funded by the Department for Communities (DfC), the course at QUB was established to address the lack of interpreters in Northern Ireland with the newly-qualified interpreters graduating next year.
Caroline Doherty is one of five deaf students taking part in the course
Deaf student Caroline Doherty, who is already fluent in both BSL and ISL, wanted to gain a formal qualification after helping friends in the deaf community who were struggling to access services.
She said deaf people have an indigenous understanding of the language which they can use to influence and make changes.
"We can share our skills, learn from that and learn more about the interpreting profession," she said.
The cohort of 17 students is a mix of BSL and ISL users, qualified and aspiring interpreters, as well as deaf and hearing students.
Ms Doherty is one of five deaf students taking part in the course.
"I can't explain how valuable this is. It's the first course where I have felt equal with my hearing colleagues."
Recently, Ms Doherty had a medical appointment that was cancelled four times because there weren't any interpreters available.
"I don't feel that's right. I can't live my life like that. I work full time. I contribute. I'm a taxpaying citizen. I should have the same rights as everybody else."
Konrad Cheng was inspired to sign up in memory of his aunt
Konrad Cheng decided to join the course in memory of his aunt who was deaf and sometimes struggled to access services.
"She went her whole life without having proper access to communication," he said.
"No one really acknowledged or remembered her. I feel like this is a personal journey in remembering her life."
At first, he found the course overwhelming due to the mix of hearing, deaf and Irish Sign Language interpreters, but said it has been a "really great learning process".
"I'm learning so much from my peers, from the qualified interpreters. I'm able to go to them and ask how they would do things in their career."
Sarah Garvey
Sarah Garvey said being a professional interpreter can be an "emotional rollercoaster"
Qualified interpreter Sarah Garvey said the profession can be an "emotional roller coaster".
"Not having enough numbers means that interpreters take on too many jobs or are in situations where there's no support," she said.
There were eight people in Ms Garvey's class when she graduated from Trinity College Dublin in 2014, of these she said only four are working in the field.
"There's a lot of different experiences, opportunities, and perspectives, but I hope as a cohort we'll be able to make a difference for the community going forward."
Communities minister introduces bill in sign language
Hospital staff ask deaf couple, 'What is a BSL interpreter?'
A federal investigator warned Maine officials about possible interpreter fraud schemes in a 2020 report that has not been disclosed until now. The findings come on the heels of fraud allegations against a major health care provider serving immigrants.
"...
In the last 10 years, providers filed more than 45,000 interpreter claims, totaling more than $41.4 million. Half of that money went to only a few providers. The BDN contacted the top 10 providers and asked if they had ever been contacted by federal law enforcement. Only two responded, with one of them saying they had not. (The BDN is not naming that provider because it has not been charged with a crime.)
The other, Gateway, said through a lawyer that it also had not been contacted in the last 10 years by federal law enforcement. The audit that prompted state officials to pause MaineCare payments to the company on Tuesday began in early 2023 and looked at claims submitted between 2021 and 2022, according to a notice of violation.
Prior to the announcement that the organization’s MaineCare payments had been suspended, the provider’s lawyer, Pawel Bincyzk, denied allegations of fraud or being aware of Pellerin’s report.
“Gateway stands by its previous statements on this issue and will continue to cooperate with the state as it has in the past,” Bincyzk said..."
"The fourth "Yunshan Cup" International Remote Interpreting Contest concluded on December 20 at Guangdong University of Foreign Studies (GDUFS), bringing together interpreters from around the world to translate local stories from Guangdong for international audiences.
More than 4,500 interpreters from China and overseas signed up for this year's competition. After multiple rounds, 251 contestants reached the finals, competing in 16 languages including English, French, Spanish, Russian, Japanese, and Portuguese. The finals were livestreamed and watched by tens of thousands of viewers online.
The competition focused on translating local stories from Guangdong for international audiences. The source materials were based on actual locations and cases from different cities, covering topics such as cultural heritage, environmental protection, industry, technological innovation, and everyday local life.
Contestants introduce Guangdong
Liu Jianda, Vice President of GDUFS, said the competition's remote format allowed young interpreters from 16 language backgrounds to compete on the same platform, making it an effective way to present China's stories to the world.
Guest speeches at the finals also sparked discussions about the future of the profession as artificial intelligence reshapes the language industry. Xing Yutang, Vice President of the Academy of Translation and Interpretation of China International Communications Group, stated that while AI tools are advancing rapidly, high-end interpreters remain irreplaceable, with their value more pronounced and demand more urgent than ever.
This view was echoed by Professor Wallace Chen Ruiqing of the Middlebury Institute of International Studies at Monterey, who argued that technology should support human interpreters, not replace them.
Sylvia Kadenyi Amisi, President of the International Association of Conference Interpreters, added that interpreters are not just language professionals but also cultural communicators, stressing that judgment, empathy, and cultural understanding remain essential to the profession, even as technology evolves.
À la CAN 2025 au Maroc, de sérieux problèmes de traduction sont constatés lors des conférences de presse.
L’utilisation d’une application de traduction inefficace et mal communiquée aux journalistes entraîne une mauvaise compréhension des échanges.
Cette situation complique le travail des médias et révèle un écart entre les ambitions annoncées pour l’événement et les réalités de son organisation sur le terrain...
"Translating Ecologies of Thought: The Digital Archive of Indigenous Language Persistence (DAILP)
22 January 2026 15:30 to 16:30
Online
Part of Computational Humanities Research Group Seminar Series
The Digital Archive of Indigenous Language Persistence (DAILP) provides a reading and writing ecology for Indigenous language practice and documentation. DAILP offers a reading and translation interface that gives indigenous peoples purpose, audience, intuitive means, and opportunity to to continue practicing their languages. The DAILP translation interface allows teams to translate language documents, audio files, and videos in archives and libraries. The translations entered into DAILP are then displayed in digital edited collections, which deepen language documentation and and encourage secondary scholarship. After a brief introduction to DAILP’s reading and writing ecologies, I will introduce DAILP’s GitHub site to share code, workflows, and technical documentation driving the reading and translation interfaces. DAILP supports the persistence and documentation of indigenous ecologies of thought.
This is an online event; the link will be communicated upon registration.
Speaker:
Ellen Cushman is Dean’s Professor of Civic Sustainability in the Department of English at Northeastern University and Co-Director of NULab for Digital Humanities and Computational Social Sciences. A citizen of the Cherokee Nation, she is project leader of the Digital Archive for Indigenous Language Persistence. Her research explores how people use language and literacy to create and endure change. DAILP has received generous support from the National Archives, the National Endowment for the Humanities, the Henry K. Luce Foundation, and the Mellon Foundation.
"(Xinhua) 09:19, December 24, 2025 JOHANNESBURG, Dec. 23 (Xinhua) -- The term "G20" was the most frequently used word in South Africa in 2025, the Pan South African Language Board (PanSALB) said on Tuesday.
PanSALB said it worked with media research company Focal Points to analyze frequency data and discovered that "G20" featured prominently in the "reputable print, broadcast and online media."
According to PanSALB, the selection process involved shortlisting candidates based on authentic language usage. "G20" emerged as the most dominant keyword largely due to South Africa's role as the G20 presidency in 2025 and its hosting of the G20 Leaders' Summit.
The terms "Government of National Unity" and "Tariffs" ranked second and third, respectively, reflecting key political developments, international engagements and socioeconomic debates that shaped the country during the year, the agency added.
PanSALB is an organization mandated to promote multilingualism, develop and preserve South Africa's 12 official languages, and protect language rights.
"Artificial intelligence-powered language translation startup Camb.ai has a new multiyear partnership with European Athletics, the governing body for sports on the continent.
The collaboration will see European Athletics use Camb.ai’s newly released publishing product, Camb.ai for Publishing, to translate its website into 12 different languages: English, Spanish, French, Italian, German, Dutch, Portuguese, Polish, Croatian, Czech, Hungarian and Serbian. Included in that work are event coverage, athlete profiles and other digital content, underpinned by AI technology that Camb.ai says can facilitate translations in more than 150 languages.
"Most tech companies treat language like a checkbox: translate the homepage, add a few help-center articles, and call it “global.”
But the next wave of growth is not coming from being “available” in more countries, it is coming from being understood, trusted, and chosen in more countries.
That is why multilingual strategy is becoming a core product and go-to-market function, not a last-minute localization task.
The problem we keep ignoring: customers don’t “mentally translate” for us
We still act like English is a universal interface, especially in SaaS and developer-first products. Yet CSA Research found that 76% of online shoppers prefer buying when product info is in their own language, and 40% will not buy from websites in other languages. If your funnel, onboarding, and support live in one language, you are not “global,” you are just easier to ignore.
The second problem: AI made translation faster, but not safer
AI translation removed the time barrier, so now everyone can ship multilingual content at speed. That sounds like a win until you realize most teams cannot confidently judge quality in languages they do not speak, especially when different AI tools produce slightly different answers. So we end up with a risky workflow: copy, paste, guess, and ship.
In 2025 and 2026, localization stops being “support” and becomes “strategy”
The localization world is already calling this out: the real advantage is integrating AI-driven localization into the broader ecosystem, so it scales with the business. When localization is wired into product, marketing, and customer success, it stops being a cost center and starts being a growth lever. If we do not build for this now, we will spend the next two years rebuilding it under pressure.
A multilingual strategy is not “translate everything”, it is “design how you scale”
It starts with picking the languages that match revenue, retention, and support demand, not just traffic. Then it defines what must be localized deeply (onboarding, pricing, trust pages, key support flows) and what can be translated lightly (long-tail docs). Finally, it sets quality rules so we get accurate translations that protect meaning, brand voice, and compliance, not just word swaps.
The real bottleneck is trust, at the sentence level
Most teams do not fail because they refuse to translate, they fail because they cannot trust what they are shipping. One bad sentence in a pricing page, security statement, or “how to cancel” flow can create churn, tickets, and legal exposure. This is why “which AI output do we believe” is now a business problem, not a linguist problem.
The solution: MachineTranslation.com’s SMART feature
MachineTranslation.com is a free AI Translator built around a simple idea: do not bet your business on one AI model’s opinion. SMART runs multiple AI engines and selects the sentence-level translation that most engines converge on, so you get one consolidated output instead of a pile of competing drafts. And because MachineTranslation.com is built by Tomedes, you can pair that AI consensus with a leading global provider known for high-quality, fast, and customized language translation services for businesses worldwide when you need expert human support.
Independent coverage of SMART describes it as consensus-based selection without an extra “rewrite layer,” which is exactly what teams need when clarity matters.
Why SMART changes the workflow, not just the output
The old workflow is messy: generate several translations, compare them manually, and hope you chose the right one. SMART flips that into a single click that surfaces where the engines agree, which is a practical proxy for reliability when you do not speak the target language. Slator also reports that internal evaluations cited in their coverage showed consensus-driven choices reduced visible AI errors and stylistic drift by roughly 18–22% versus relying on a single engine, which is the kind of improvement that compounds at scale.
Where this matters most for tech businesses: growth content
Multilingual SEO is exploding because AI can generate and translate at volume, but volume does not help if the translation bends intent or weakens a call-to-action. SMART is built for exactly that moment when you need speed and confidence at the same time, especially for landing pages, app store listings, and lifecycle emails. If we want professional translations without slowing the team down, consensus beats guessing every time.
Where this matters most: product and support
Your UI strings, onboarding tooltips, and help articles must be consistent, or users feel friction even if the translation is technically “correct.” SMART is designed to reduce hallucinations and outlier phrasing by favoring the majority over the weird one-off output, which helps keep terminology stable across releases. That means fewer support tickets caused by language confusion, and faster localization updates when you ship new features.
The part many teams forget: security and governance
Translation often includes contracts, customer data snippets, incident notes, and internal docs, so “free online translator” is a real risk. MachineTranslation.com documents Secure options, including a secure translation mode processed on private servers and a Secure Mode approach that restricts processing to SOC 2-compliant sources. That matters because multilingual strategy is also data strategy, and trust is hard to rebuild once you lose it.
A future-focused take: consensus is how we survive the model explosion
We are heading into a world with more models, more outputs, and more “almost right” translations that look fine at a glance. Consensus-based translation is a sane response to that future because it reduces dependence on any single system and gives teams a clearer baseline for review. Even MultiLingual’s recent coverage ties reliability to “more data points,” noting that SMART uses algorithmic voting to pick the best sentence-level translation.
What you can do this quarter, without boiling the ocean
Start by mapping your revenue funnel and support journey, then pick 2–3 languages where better understanding would most directly lift conversion or reduce tickets. Next, standardize your translation workflow so marketing, product, and support are not using different tools and creating inconsistent language. Then operationalize SMART as the default for first-pass production content, and reserve human review for the truly high-stakes pages where you want maximum assurance.
The bottom line
A multilingual strategy is not about sounding international, it is about building a business that can scale trust across borders. If your team is shipping faster than it can verify, you do not have a growth engine, you have a risk engine. MachineTranslation.com’s SMART feature is a practical way to turn multilingual growth into something you can actually run, measure, and trust.
"La traduction littéraire est un art de la passion et de la précision, qui comporte son lot de défis à l’heure actuelle.
Publié à 8h17
Sarah-Louise Pelletier-Morin
Quand Arianne Des Rochers tombe sur This Little Art, l’essai de Kate Briggs, c’est le coup de foudre. « Ce livre-là m’a complètement happée… je me suis dit que j’aimerais vraiment le traduire et que ce serait le projet d’une vie », raconte-t-elle. Problème : Briggs est britannique. Au Canada, les principales subventions de traduction ne couvrent que les auteurs canadiens. Traduire Le petit art ne serait donc pas financé.
Après que la traductrice eut cogné à la porte de plusieurs maisons d’édition, Le Quartanier finit par accepter le projet. Un « projet de passion » qui devient ainsi en novembre dernier un livre disponible en français — au prix, entre autres, d’un travail quasi bénévole, rendu possible parce que Des Rochers vit surtout de l’enseignement et de la recherche.
Titulaire de la Chaire de recherche du Canada en traduction et colonialisme depuis juillet 2024, Arianne Des Rochers s’est spécialisée dans les littératures autochtones et queers. Pour traduire des œuvres autochtones, elle travaille en étroite collaboration avec les auteurs et des partenaires pour « combler le fossé » entre ses expériences et celles des textes qu’elle traduit. Pour elle, traduire n’est pas corriger une « copie conforme » de l’original, mais inventer une relation : « Lire une traduction, c’est lire un livre dans les yeux de quelqu’un d’autre. Le petit art, c’est ma façon à moi de raconter le livre de Kate Briggs à d’autres gens. »
Savoir en mouvement Cette idée de la traduction comme savoir en mouvement traverse l’essai de Briggs. Plutôt que de prouver une expertise figée, « la traduction produit de nouveaux savoirs sur le monde ». Elle devient « un site où apprendre, par la lecture et l’écriture », un lieu d’expérimentation qui invite « de plus en plus de gens à traduire ». Loin de l’obsession de la « faute », Briggs insiste sur le caractère spéculatif de la traduction : on demande au lectorat d’accepter qu’un texte pensé dans une autre langue existe en français ou en anglais, de suspendre encore un peu plus son incrédulité. Il y a là, dit-elle, « quelque chose de spéculatif, voire de romanesque ».
À Gatineau, Madeleine Stratford parle, elle aussi, de la traduction comme d’un geste de lecture poussée à l’extrême. Poète, traductrice et professeure à l’Université du Québec en Outaouais, elle vient de la littérature allemande et espagnole, puis d’un mémoire où l’on exige qu’elle traduise son propre corpus. « J’aime dire qu’il faut déconstruire le texte original pour le reconstruire avec des matériaux différents. Mettons que j’ai une maison en bois, mais que je n’ai que de la brique pour la rebâtir : il faut que ça ait l’air d’être du bois », image-t-elle.
Comme beaucoup de traductrices littéraires, Stratford n’a pas de formation pratique formelle, mais une longue fréquentation de la poésie et du roman. Les meilleurs traducteurs, selon elle, sont « d’excellents lecteurs et lectrices », capables de disséquer un style, un rythme, un ton pour mieux les recréer ailleurs. Elle-même a signé des dizaines de traductions — poésie, fiction, essai, jeunesse — tout en apprenant « à écrire » en se glissant dans les voix des autres.
Son regard est sévère envers une culture qui s’intéresse aux traductions surtout pour y traquer des erreurs. « Quand on parle de traduction, on va chercher toutes les petites bêtes noires et dire : “regarde comment c’est mal fait” », déplore-t-elle. Or, une traduction est « le fruit d’un travail collectif et d’une négociation constante » avec l’éditeur, le réviseur, parfois l’auteur, et elle n’a pas pour but d’effacer la distance avec l’original : elle la met en scène, l’assume.
La forme la plus assidue de lecture Dans L’envers de la tapisserie, un essai d’Alberto Manguel qui paraîtra sous peu en traduction française, le lecteur est invité à regarder ce qui se trame au dos du texte. « La traduction peut être (doit être) la forme la plus assidue de lecture », écrit-il, rappelant que le traducteur s’adresse avant tout au lecteur de la traduction. Le mythe d’Orphée lui sert de métaphore : Orphée, traducteur d’Eurydice, perd son amour lorsqu’il se retourne pour vérifier qu’elle le suit. « Le miracle de la traduction est un acte de résurrection » : il faut accepter de laisser l’original dans son royaume pour que la version traduite vive pleinement. L’art de traduire rappelle ainsi « qu’il n’existe jamais de lecture “exacte” » : Balzac lu par Freud n’est pas Balzac lu par Marx, et Balzac traduit n’est jamais exactement Balzac.
À Berlin, Jennifer Drummer incarne une autre facette de ce « petit art ». Germanophone tombée amoureuse du Québec lors d’un premier hiver glacial en 2008, elle y revient, étudie à l’Université de Montréal, puis fonde une entreprise de promotion de la littérature et de la musique québécoises dans l’espace germanophone. Sur ses blogues, dans ses événements et ses traductions, elle s’emploie à « ramener dans [sa] culture » des œuvres encore inédites en allemand.
Passée par le programme franco-allemand Georges-Arthur Goldschmidt, elle cotraduit des romans et des essais québécois et signe la version allemande de La bête à sa mère de David Goudreault, la poésie de Joséphine Bacon et un ouvrage de vulgarisation scientifique sur le cerveau et la musique. Comme ses collègues, elle ne vit pas de la traduction : elle cumule mandats de communication, animation, interprétation, résidences de création. En Allemagne, raconte-t-elle, l’association des traducteurs est très active. La campagne Name the translator pousse médias et maisons d’édition à faire apparaître les noms partout, y compris sur les pages couverture.
Défis actuels et futurs Toutes trois, pourtant, se heurtent à des conditions matérielles fragiles. Stratford évoque ces albums jeunesse étrangers traduits pour « environ 500 dollars » — des semaines de travail pour des sommes faméliques. Elle note aussi qu’il y a aujourd’hui « beaucoup moins de contrats » qu’avant, en partie à cause des changements récents dans les programmes du Conseil des arts : disparition d’un taux plancher, enveloppes stagnantes, ce qui encourage certaines maisons à mieux payer quelques projets… au prix d’en traduire moins. Drummer, de son côté, peut accepter des projets « hors normes » parce qu’elle a d’autres sources de revenus.
La menace de l’intelligence artificielle plane sur cet équilibre déjà précaire. Pour Stratford, l’IA dite générative n’est qu’« une intelligence de corpus », purement mathématique, incapable de ressentir. Or, en traduction littéraire, « le ressenti est parfois plus important que l’intelligence » : il faut décoder un ton, un effet, oser s’éloigner du texte source pour être fidèle à son impact. « L’ordinateur va donner dans la probabilité ; nous, on peut créer du neuf », soutient-elle. Drummer partage cette méfiance, parlant carrément de « vol » quand des entreprises entraînent leurs modèles sur des corpus littéraires sans demander le consentement des auteurs et des traducteurs ni les rémunérer.
Malgré tout, le désir demeure au cœur du métier. Désir d’un texte, d’une voix, d’un monde étranger dans lequel on choisit de « baigner » pendant des mois, comme le dit Arianne Des Rochers. Désir d’apprendre en lisant et en écrivant, comme le formule Kate Briggs. Désir, enfin, de lecteurs, ceux pour qui on travaille dans l’ombre, en acceptant que l’original se perde un peu pour que la traduction existe. Peut-être est-ce là, justement, ce « petit art » dont parlent Briggs et Manguel : une pratique du détail, de l’attention, mais aussi une manière très concrète d’élargir nos bibliothèques et nos horizons. Encore faut-il que ces passeuses de langue puissent, elles aussi, être vues, lues et reconnues." https://www.ledevoir.com/lire/944031/traduction-est-surtout-pas-petit-art #Metaglossia #metaglossia_mundus
"Des missionnaires chrétiens veulent traduire la Bible dans près de 7 000 langues avant 2033. Leur usage de l’intelligence artificielle soulève des questions théologiques et révèle les limites des modèles de langage actuels, observe “The Economist”.
À cette période de l’année, des centaines de millions de personnes se rendent à l’église. Même les brebis égarées du chemin du Seigneur retournent souvent au bercail à Noël. Elles viennent écouter l’histoire de Marie, Joseph et l’enfant Jésus installé dans une mangeoire, car il n’y avait plus de place pour eux à l’hôtel
Les missionnaires chrétiens sont persuadés que davantage d’âmes pourraient être sauvées par l’histoire du Christ. Disponible dans plus de 750 langues, la Bible est déjà le livre le plus traduit au monde, mais ils souhaiteraient qu’elle le soit dans chacune des près de 7 000 langues vivantes de la planète. Pour atteindre cet objectif, les croyants ont fait appel à un nouvel outil dans leur quête spirituelle : l’intelligence artificielle (IA).
La traduction est un chemin de croix. L’Ancien Testament compte environ 600 000 mots et on raconte que la traduction de chacun d’eux aurait nécessité le concours de 70 savants au IIIe siècle avant J.-C. Le Nouveau Testament, lui, est rédigé en mauvais grec, ce qui n’aide pas. La majeure partie reste ambiguë : personne ne sait par exemple ce que signifie epiousion dans la phrase “Donne-nous aujourd’hui notre pain epiousion”, mais les traducteurs ont opté pour “quotidien”..." The Economist Traduit de l’anglais Publié le 24 décembre 2025 Lire la suite 👇🏿👇🏿👇🏿 https://www.courrierinternational.com/article/religion-traduire-la-bible-dans-l-ensemble-des-7-000-langues-un-miracle-nomme-ia_238506 #Metaglossia #metaglossia_mundus
"Nombre de traducteurs redoutent un déferlement de l’intelligence artificielle de plus en plus utilisée par les éditeurs, qui ne la mentionnent pas toujours. Au risque de voir des milliers d’emplois disparaître, et que seule une poignée d’ouvrages littéraires ou de sciences humaines soient, à l’avenir, encore traduits par des humains.
Par Nicole Vulser (Arles [Bouches-du-Rhône}, envoyée spéciale) Publié hier à 05h45, modifié hier à 13h09
« Je n’aurais jamais cru que, de mon vivant, je verrais disparaître mon métier. C’est pourtant ce qu’il se passe », se désole la traductrice anglais-russe-français Karine Reignier-Guerre, âgée de 55 ans. Chargée de cours et tutrice en master de traduction littéraire à l’université Paris Cité et à l’université d’Avignon, elle propose aujourd’hui à ses étudiants une approche très concrète de l’intelligence artificielle (IA) générative. « Si c’est un cataclysme, il faut leur en parler et les préparer à l’affronter », dit-elle. Chaque année, depuis 2020, elle choisit donc un ouvrage de cosy mystery, un sous-genre de fiction policière sans violence, se déroulant souvent dans les confins de la campagne anglaise, à l’écriture à la fois simple et très codifiée, dialoguée et pauvre en métaphores. « Je propose deux feuillets aux étudiants de master 2 en fin de cursus. Ils traduisent le texte et j’anonymise leurs versions. J’utilise aussi un texte traduit par un logiciel comme DeepL », détaille-t-elle. A l’aveugle, les étudiants doivent ensuite distinguer la version humaine rédigée par leur voisin et celle réalisée par la machine. « Il y a trois ans encore, tous faisaient immédiatement la différence. Aujourd’hui, presque tous se trompent : les marqueurs qui trahissent l’IA sont devenus bien plus difficiles à déceler »..."
"Adjunct Assistant Professor and Writing alum Miranda Mazariegos '25 has been awarded a 2025 Granum Foundation Translation Prize, awarded each year to support US-based translators as they complete a work of translation into English.
The award was given to Mazariegos for her translation of a collection of three short novellas by Arnoldo Gálvez Suárez titled Alguien bailará con nuestras momias. Mazariegos shared, "Winning the Granum Foundation Translation Prize last month was such an honor! Arnoldo Gálvez Suárez is a Guatemalan contemporary author and his work caught my attention because of its voice and imagery; his long-winded sentences capture details that portray the essence of Guatemala, where I am also from. So, translating gave me an opportunity to feel closer to my home, to imagine it and enjoy it from afar, as I pursued my MFA in New York."
Her time at Columbia greatly shaped the finished product that has now received accolades. "This project was a part of my literary translation thesis at Columbia, so I am especially thrilled to see how it was not only my work, but also the feedback of my peers and professors, that helped me get this award. Practically speaking, the prize is also huge, as I'm sure it will help me as I look for a publishing home for this wonderful little book."
Mazariegos is a writer, editor, and literary translator originally from Guatemala City. She began her career in radio, working in various roles for NPR shows such as Radio Ambulante, Book of the Day, Throughline, and Weekend All Things Considered. Her work, which covers Latin America's art, culture, and politics, has been published by Americas Quarterly, NPR, VICE News, and KPCC, among others. She translates both from and into Spanish, and her translations have been published or supported by World Literature Today, Asymptote Magazine, the Bread Loaf Writers’ Conferences, and the British Centre for Literary Translation (BCLT). She holds an MFA from Columbia University, where she now teaches Undergraduate Nonfiction. She is an editor at Americas Quarterly and lives in New York City." https://arts.columbia.edu/news/adjunct-assistant-professor-miranda-mazariegos-25-wins-granum-foundation-translation-prize #Metaglossia #metaglossia_mundus
"The promise of technology as a saviour isn’t entirely wrong. But it will remain a myth until it serves us in the languages we live in, speak in, and dream in. Khansa Maria Published December 22, 2025 Updated a day ago
Technology loves to promise miracles. For people like me, it is often marketed as a great equaliser; an assurance that with the right device, disability is not a barrier. Access is just a download away, we are told.
But growing up, I realised that this flashy narrative, wrapped in lofty guarantees, was merely a mirage — out of reach for me.
For a family like mine, buying tech was expensive. My mother worked tirelessly to purchase devices she hoped would level the field for my visually impaired brother and me. But even when she managed to get them, a more invisible barrier persisted: hardly any technology could speak our tongue.
A profound disconnect While I could listen to English books through screen readers, reading or writing in Urdu — the language of home, of emotion, and of identity — remained a distant dream. I longed to read novels on my own, to write my own stories, to revise schoolwork without leaning on someone else’s eyes.
That dream became especially far out of reach during O-levels, when I was barred from sitting the Urdu exam simply because no accessible technology existed to support the process.
Urdu is compulsory for university admissions in Pakistan, and, like many visually impaired students, I found myself locked out of a gateway to higher education. Not because I lacked talent, but because the tools did not exist.
Today, decades later, much has improved, and for that I am grateful. But the truth is that progress has been slow, inconsistent, and unfairly distributed. My research with visually impaired communities across Pakistan reveals that the very barriers I encountered at that time continue to persist to date.
Many participants described a profound disconnect from technology, particularly those without English proficiency. We often assume that access to a smartphone equals access to the world. For some of us, that assumption remains painfully untrue.
Across Pakistan, Punjabi, Pashto, Sindhi, Balochi, Saraiki, and dozens of other languages are spoken every day. The country is home to at least 80 languages and nearly 27 million people with disabilities, many of whom depend on technology for education, employment, and basic communication.
When the tools they rely on cannot understand their language, the consequence is not mere inconvenience. It is exclusion: systemic, silent, and on a massive scale.
The myth of inclusive technology What does that look like day to day? The accessibility audits carried out in the Accessibility, Language, and Tech for the People (ALT) project paint the picture clearly.
Starting with screen readers: on Windows, the commonly used E-Speak Urdu TTS engine performs so poorly that even the simplest tasks become laborious. Several participants said the voice was so robotic and unclear that reading long-form content became overwhelmingly difficult.
Starting with screen readers, on Windows, the commonly used E-Speak Urdu TTS engine performs so poorly that even the simplest tasks become laborious. Several participants noted that the voice was so robotic and unclear that reading long-form content became extremely difficult.
On Android, the picture is slightly better, and Google’s speech services provide clearer Urdu pronunciation. But typing remains an obstacle even then because the system does not read individual letters aloud, and therefore, users are unable to confirm what they have written. Yet another obstacle.
On iPhones, things get even more complicated. There is no built-in Urdu TTS at all.
This leaves visually impaired users with no option but to change system settings manually every time they want a message to be read in Urdu, or worse, have to rely on clumsy third-party tools. One participant put it starkly: writing even a single sentence in Urdu using the available synthesisers was “a terrible experience” and felt nothing like the smooth, accessible typing they were used to in English.
Beyond language itself, application accessibility is equally inconsistent. People often assume that because an app works perfectly for sighted users, it must work for everyone. But the audits show a very different reality.
Daraz, one of Pakistan’s most widely used shopping platforms, remains deeply problematic. Product images lack descriptions. Buttons are unlabelled. Tab navigation is inconsistent. Product details are sometimes impossible to access without help. Checkout processes fail to provide feedback, leaving users unsure whether an action has been completed. Several testers described having to rely on sighted assistance or apps like “Be My Eyes” just to finish an order.
Ride-hailing apps mirror these challenges. InDrive is relatively better, but nearly all such apps fail in one crucial area: live location tracking. For visually impaired passengers, this isn’t a bonus feature; it’s essential for safety and independent travel. Yango, for instance, makes even entering a location difficult because of inconsistent labelling. Bykea has similar navigation issues.
Food delivery apps, on the other hand, are a mixed bag. Foodpanda is usable for many tasks, but placing an order or chatting with a rider can become difficult due to unlabelled options or delayed keyboard feedback. Some testers could search for restaurants easily, only to find themselves blocked at the final step because the “place order” button wouldn’t register with their screen reader.
Banking apps, perhaps the most essential tools of modern life, show the widest range.
SadaPay is frequently praised for its smooth and intuitive accessibility. But Easypaisa and JazzCash — two financial lifelines for millions — can be nearly impossible to navigate. Users described crowded interfaces, unlabelled buttons and processes so convoluted that beginners “give up out of frustration and demotivation”.
Think about what that means: in 2025, millions of blind or visually impaired users still cannot send money, order food, track a ride, or buy groceries without assistance. We often talk about accessibility as if it’s a checklist — add alt text here, label a button there, job done.
But accessibility isn’t a feature. It’s a commitment. A philosophy. A willingness to build with the most marginalised users in mind, not as an afterthought but as a starting point. Right now, the digital world is being shaped around English-speaking, sighted, neurotypical users. Everyone else is left to catch up.
Redefining accessibility For me, participating in the ALT project helped reclaim some of the marginalisation I grew up with. The project examines how language politics, disability and technology intersect in South Asia. Though these countries share deep histories and cultures, we found that barriers vary, and that solutions must be community-driven, co-created and grounded in the lived realities of disabled people.
Time and again, participants expressed a desire to separate themselves from colonial linguistic hierarchies. They don’t want to centre English; they want to centre their languages — to read the news in Pashto, to type in Punjabi, to browse a website in Sindhi, to write poetry, read novels and order food in Urdu without fighting their phone every step of the way.
And yet, big tech companies have been slow to respond. Apple still doesn’t offer Urdu TTS support. Android has made improvements, but it remains far from ideal. Community whispers about “something new coming soon” have circulated for years, only to fade away as researchers hit barriers, lose funding or abandon the project. The cycle repeats.
Pakistan deserves better. South Asia deserves better. Every language deserves digital visibility, not only the global ones. And most of all, every disabled person deserves the dignity of independent access.
Accessibility, in its current form, has too narrow a definition. It must widen linguistically, culturally and technologically. Because when technology fails people who need it most, it isn’t just a technical issue — it’s a question of justice, dignity and belonging.
The promise of technology as a saviour isn’t entirely wrong. But it will remain a myth until it serves us in the languages we live in, speak in, and dream in." https://www.dawn.com/news/1960708 #Metaglossia #metaglossia_mundus
"Geoffrey Chaucer is known as the "Father of English Literature" or the "Father of English Poetry" for his medieval classic "The Canterbury Tales," a work that encouraged writers of his time to write in Middle English rather than French.
Over six centuries later, Francis So (蘇其康) released in October the first complete Chinese translation of "The Canterbury Tales" by a Taiwanese translator.
In an interview with CNA in late November, So said Chaucer wrote during a period when French still dominated literary culture.
The publication of "The Canterbury Tales" helped popularize Middle English, while its poetic techniques shaped later writers, including Shakespeare, he said.
So hopes the new edition will inspire more young researchers to build and carry forward Taiwan's tradition of medieval Western literary studies.
Reliving the pilgrimage So noted that "The Canterbury Tales," written in the late 14th century, depicts a pilgrimage of 30 Christians traveling from London to Canterbury to venerate St. Thomas Becket.
The pilgrims take turns telling stories along the way, forming the work's narrative frame.
Although Chaucer originally planned 120 tales -- two for each pilgrim on both the outward and return journeys -- only 24 survive, preserved mainly in two manuscripts.
So based his translation primarily on the more complete Ellesmere manuscript and consulted the Hengwrt manuscript, which scholars believe reflects the earlier state of Chaucer's text.
So said he adopted "fidelity" as his guiding principle, preserving original syntax, word order and imagery whenever possible.
"If the original uses a noun, I try to translate it as a noun. Sometimes reversing the sentence order makes the Chinese more fluent, but it weakens fidelity to the text," he said.
To help contemporary Taiwanese readers navigate the unfamiliar medieval world, So included extensive annotations, particularly on material culture and institutional structures -- a key feature distinguishing his version from the earlier translations.
So credits his sensitivity to historical and cultural nuance to the rigorous comparative-literature training he received in the United States, where he studied multiple languages and took courses in translation studies.
Born in 1948, So earned his bachelor's degree at National Taiwan University's Department of Foreign Languages and Literatures and later received a Ph.D. in comparative literature from the University of Washington in Seattle.
He is currently an emeritus professor at National Sun Yat-sen University, where he has taught since 1983.
So said translation involves far more than "looking up words in a dictionary."
To better reconstruct the medieval pilgrimage, he visited the British Museum in 2023 to consult historical materials and traveled portions of the route described in the text.
This fieldwork, he said, helped him handle place names and cultural references with greater accuracy.
Tradition and legacy So said "The Canterbury Tales" continues to resonate today, noting that contemporary British writer Zadie Smith drew inspiration from "The Wife of Bath's Tale" for her play "The Wife of Willesden."
Elsewhere, J.K. Rowling has also acknowledged that "The Tale of the Three Brothers" in "Harry Potter and the Deathly Hallows" is an allusion to "The Pardoner's Tale."
Reflecting on his academic career in Taiwan, So said there were no scholars specializing in medieval literature when he studied at NTU, and he resolved to help establish the discipline when he undertook graduate studies in the U.S.
"When we founded the Taiwan Association of Classical, Medieval and Renaissance Studies (TACMRS) in 2007, that meant the establishment of a tradition," he said.
The association's annual conference has since become a key event for domestic scholars and students in the field.
Nearly two decades after the founding of TACMRS, So said he is heartened to see more emerging scholars entering the field.
He hopes the new translation will lower the barrier of entry for readers and encourage more people to pursue medieval literary studies." INTERVIEW/New translation of 'The Canterbury Tales' marks milestone in Taiwan medieval studies 12/22/2025 12:11 PM By Chao Yen-hsiang, CNA staff reporter https://focustaiwan.tw/culture/202512220006 #Metaglossia #metaglossia_mundus
"MUSCAT: The Bait AlGhasham–DarArab International Translation Prize on Sunday announced the shortlists for its 2026 cycle across three categories — Translators, Authors and Omani Publications — following the completion of all judging stages and final reviews. The 2026 shortlists reflect a cycle marked by wide participation and notable diversity, spanning multiple literary genres and a broad range of geographic backgrounds. This year also saw submissions from translators based outside the Arab world, underscoring the Prize’s expanding regional and international reach. Shortlist – Translators Category in alphabetical order by English title Angel of the South, translated by Peter Theroux — by Najem Wali Cairo Marquette, translated by Katherine Van de Vate — by Tareq Imam People and Lizards, translated by Osama Hammad and Marianne Dhenin — by Hassan Abdel Mawgoud Things Are Not in Their Place, translated by Zia Ahmed — by Huda Hamad Village of the Hundred, translated by Enas El Torky — by Rehab Luay Shortlist – Authors category, alphabetical order by Arabic title Abad Ghayr Mariyya — Short-story collection by Mustafa Mallah Al Matador — Novel by Majdi Daibes Jibal Al Judari — Novel by Abd Al Hadi Shaalan Jarash Jarash — Novel by Wael Raddad Kain Ghayr Sawi — Novel by Tahir Al Noor Shortlist – Omani publications category, alphabetical order by Arabic title Tahta Zill Al Zilal — Novel by Mohammed Qart Al Jazmi Arous Al Gharqa — Novel by Amal Abdullah Qawanin Al Faqd — Short-story collection by Mazen Habeeb The shortlists were selected by independent judging panels operating under the supervision of the Prize’s Board of Trustees, chaired by Marilyn Booth, an internationally recognised translator and scholar of Arabic literature. Board members also include Mohammed al Yahyaei, a writer and cultural historian, and Sawad Hussain, an award-winning Arabic-to-English translator. The Translators Category jury comprised of Dr Samaher al Dhamen, a specialist in comparative literature and cultural studies; Marcia Lynx Qualey, founder of ArabLit and a leading advocate of Arabic literature in translation; and Dr Luke Leafgren, an Arabic-English literary translator and two-time recipient of the Saif Ghobash Banipal Prize. The Authors and Omani Publications categories were judged by Dr Amir Taj Al Sir, a widely translated novelist and physician; Bushra Khalfan, a prominent voice in contemporary Omani literature; and Yas al Saeedi, an Iraqi poet, novelist and playwright whose work has garnered multiple Arab and international awards. According to official figures, the 2026 cycle received 346 submissions in the Authors Category, 34 in the Translators Category and 10 in the Omani Publications Category. Launched three years ago through a partnership between the Bait AlGhasham Foundation for Press, Publishing and Advertising and DarArab for Publishing & Translation, the Prize is administered by DarArab and funded by the Bait AlGhasham Foundation. It aims to support Arabic literature and expand its global circulation through translation and international publishing. The Prize carries a fund of £70,000 (approx), allocated across financial awards and professional support for translation, editing, publishing and international promotion. The winning works will be announced in conjunction with the Muscat International Book Fair 2026." Bait AlGhasham–DarArab Translation Prize unveils 2026 shortlists https://share.google/sF1u9XSkllOvVs5kX #Metaglossia #metaglossia_mundus
To get content containing either thought or leadership enter:
To get content containing both thought and leadership enter:
To get content containing the expression thought leadership enter:
You can enter several keywords and you can refine them whenever you want. Our suggestion engine uses more signals but entering a few keywords here will rapidly give you great content to curate.
Your new post is loading...
Scoop.it!
