I’m trying to get Coqui AI’s TTS to speak out phonemes. The input can be either IPA or Arpabet – it doesn’t really matter. I can’t see anything in the docs that points to how to do this but it certainly feels possible – is it?

If it’s relevant I’m using tts_models/en/jenny/jenny but I’m not bound to that model.

If this isn’t possible using Coqui-AI then is an alternative available which can speak out phonemes? I’ve tried eSpeak but haven’t been able to get it to work (I tried by invoking the tool on the command line but it doesn’t produce an output). This is on Windows. I also understand that the voice quality of eSpeak isn’t great.

Thank you