Open qaissikandar opened 6 months ago
For both cases you just need a ws server fot stt and a http server for tts. Let say you have whisper in local for tts, at the end of the day you just need to expose and API that recieve a voice and a text. Then just use a tunneling tool to get a production ready url.
The new Jambonz has the ability to use custom services for STT and TTS. let's say I have deployed my model, how can I configure it?