Google Updates Translate with AI-Powered Context and Interactive Features
Google rolls out AI-powered updates to Translate that introduce contextual explanations, alternative phrasing, and interactive...
Maria Stasimioti
Array
(
[post_type] => post
[posts_per_page] => 12
[paged] => 4
[post_status] => publish
[author] => 2795
)
OECD Launches RFPs for Translation Services and Cloud-Based TMS
The OECD launches separate tenders for multi-year translation, revision, and post-editing services and a cloud-based...
European Commission Seeks New Boss for Translation Centre
The European Commission has opened applications for the next Director of the Translation Centre for...
Researchers Aim to Make Human Translation Evaluation Easier to Run
A new platform from ETH Zurich and Cohere researchers aims to make structured human evaluation...
AI Translation Evaluation Moves From LLM-as-a-Judge to Agent-as-a-Judge
AI evaluation is moving beyond LLM-as-a-judge toward Agent-as-a-Judge, with translation explicitly in scope — solid...
Academia and Hyperscalers Building the Core Infrastructure for African Language AI
Recent announcements from academia, Google, and Microsoft highlight ongoing efforts to expand AI support for...
Reddit Ties AI Translation to FY2025 User Growth
In its annual report, the US-based social media and community platform Reddit highlights AI translation...
Prompt-Based Control Reaches Enterprise Speech-to-Text
Recent releases from Microsoft and AssemblyAI reflect growing interest in structured, configurable speech recognition as...
France’s National Tech Lab Questions Reliance on Prompts in AI Translation
An Inria study suggests that internal control mechanisms in large language models may reduce reliance...
What Text-to-Speech Evaluation Misses in Real-World Deployment
Researchers argue that text-to-speech evaluation under-tests what matters in real-world deployment — and propose a...
Microsoft Unveils VibeVoice-ASR for Long-Form, Multi-Speaker Transcription
Microsoft releases VibeVoice-ASR, an open-source speech-to-text model designed for long-form audio, structured transcription, and customised...