Azure Speech LLM 2607: Better Multilingual Transcription in Microsoft Foundry
Azure Speech LLM 2607 is here, with more accurate speech-to-text for multilingual and mixed-language audio. In this quick demo, we compare LLM Speech 2607 with the previous 2605 model in the Microsoft Foundry Speech-to-Text playground. You'll see: ✅ More accurate recognition when speakers switch languages mid-conversation ✅ Improved number handling ✅ Better transcripts for voice agents, contact centers, meeting transcription, and multilingual apps, with no changes to your app architecture Try LLM Speech 2607 today in Microsoft Foundry, or build with the Speech REST APIs and SDKs. Learn more: h
Discussion
Rejoignez la communauté pour réagir et poser vos questions.
Aucun message pour l'instant. Lancez la discussion !