I've integrated locally speech to text and text to speech in an optional way to my instance using whisper and kokoro on docker locally.
If I send an audio to the chat, it gets transcripted, shown on the conversation and sent to claude.
If claude answers a long message I can reply "read" to that message and The long message will get read and sent via an audio note.
Let me know if you find this useful and I can review it all and open a feature branch. Regards and thanks for your project.
I've integrated locally speech to text and text to speech in an optional way to my instance using whisper and kokoro on docker locally.
If I send an audio to the chat, it gets transcripted, shown on the conversation and sent to claude.
If claude answers a long message I can reply "read" to that message and The long message will get read and sent via an audio note.
Let me know if you find this useful and I can review it all and open a feature branch. Regards and thanks for your project.