Sotto currently uses Kokoro as its local voice backend. It would be useful to explore a model picker so users can choose other local TTS models when they are practical on macOS.
Goal:
- Investigate alternate local models such as PocketTTS
- Identify what model metadata/settings Sotto needs to switch between local models
- sketch a model picker UX in Sotto Studio and keep Kokoro as the default model.
- Avoid coupling the app too tightly to one backend shape.
This is a good first issue if scoped to research, interface design, and a small proof of concept rather than full production support.
Sotto currently uses Kokoro as its local voice backend. It would be useful to explore a model picker so users can choose other local TTS models when they are practical on macOS.
Goal:
This is a good first issue if scoped to research, interface design, and a small proof of concept rather than full production support.