streaming support
#1
by Dusk007 - opened
Does this model supports streaming for realtime voice agents applications on offline or paid service ?
Hi! Thanks for your question. This repository currently doesn't expose a dedicated real-time streaming API. There is an open-source example of using our model in a real-time transcription setting: mlx-audio-swift, along with a demo here. If you're looking to build a real-time voice agent, that repository would be a good starting point.
For deployment, you have two options:
- Offline/self-hosted: download the model weights from HF and run it locally;
- Use the Cohere API (free for experimentation subject to rate limits, with production options available).
shaun-cassini-cohere changed discussion status to closed