Whisper support for Speech to Text and Elevenlabs support for Text to Speech

[maddyonline]

This is a great package. I am wondering if we want to implement a custom Transcriber (say based on OpenAI's whisper) how would we go about doing it? Any code snippets (even if incomplete) would be useful. In particular, how to convert incoming packets compatible with Whisper?

Likewise if we want to use a different Synthesizer like ElevenLabs what is a good place to start.

I would be happy to contribute these additions to the package if I can get some help on getting started.

[lmsutools]

I suscribe to this idea. Also, ElevenLabs with stream option to receive faster audios.