Cohere launches an open source voice model specifically for transcription

Relatively light at just 2 billion parameters, the model is meant for use with consumer-grade GPUs for those who want to self-host it. It currently supports 14 languages.

Fonte: TechCrunch
Leia mais: https://techcrunch.com/2026/03/26/cohere-launches-an-open-source-voice-model-specifically-for-transcription/