Feature Request: Upgrade to Coqui voice stack

[JRMeyer]

👋 Hi there

A lot has changed since my last PR was merged here. It's still an awesome project, but the quality of STT and TTS has greatly improved since the (now unmaintained) mozilla repos were used.

It would be very easy to upgrade to Coqui (🐸STT and 🐸TTS) and get those improvements:)

[sara-tagger]

Thanks for raising this issue, @yennycheung will get back to you about it soon✨

Please also check out the docs and the forum in case your issue was raised there too 🤗

[AshutoshDongare]

Please check out https://github.com/AshutoshDongare/convo which integrates Silero speech models with Rasa for voice interface.

[gullabi]

It seems this repository is no longer maintained. I am working on integrating 🐸TTS to the voice interface as @JRMeyer requested. Just popping in here to give some details to whomever is interested.

The linked blog post in the README file does not include the socketio connections for speech tools (the code snippets are not copied from the original post), but the correct reference is the medium article and hence the file to be updated is in this gist.

In principle, I will be working on this these weeks, but if anyone else is interested in lending a helping hand, let me know.