Mistral Voxtral 3 replaces three separate vendor contracts with one voice agent SDK, outperforming Whisper large-v3 on ...

For some time now, I have been comparing the terms of service for AI transcription services. ChatGPT, Gemini, Copilot, Slack, ...

I believe many executives use the AI voice recorder "Plaud." You are able to record. Transcriptions and summaries are ...

Mistral AI launches Voxtral 3 with speech-to-text, TTS, and a voice agents SDK, with API pricing expected near $0.001 per minute.

Revenue has an imperfect relationship with reality at the best of times. It can be booked long before cash arrives, or long ...

Mistral says its new Voxtral Small Transcribe 3 hits 4.1% word error rate on the FLEURS 45 multilingual benchmark when the language is specified, and ...

A bidirectional AI model combining speech acoustics with transcript meaning reached 90.42% accuracy on the ADReSS Alzheimer’s disease benchmark in a small retrospective study.

Handy replaced my $10-a-month transcription app, and it actually understands how I talk.

Conversational AI and voice large language models are reshaping how we understand empathy and change the nature of work ...

A new study shows that synthetic speech generated by large language models significantly improves speech recognition for ...

An internal model trained from August 28 has reportedly resolved the Navier-Stokes equations and more than 100 long-standing open problems. No public proofs have been released.

Typing long documents, emails, or notes on a computer can take time, especially when you have a lot to write. Voice dictation offers a convenient alternative, but many speech-to-text tools rely on ...