Local Inference

[pavankalyan066]

I am interested to run this in local and send my stream to the model to and see the output in the command line.

Do we have possibility for that, I mean the way how we send audio bytes to whisper or any asr model.

Thank you.

[Paulmzr]

@pavankalyan066 Hi, thanks for your attention! We are actually now plan to develop utils to deploy our model in local. Currently the released code is only for the evaluation using SimulEval.

In my opinion, the framework of the utils is clear: Receive the speech signal every chunk size and call the agent. I am planing to do this recently and I may use the code in Seamless Streaming for reference.

If you have and ideas or are willing to contribute the code, don't hesitate to tell me. :)