Could we please get some sort of Whisper/Gemini Flash/speech-to-text integration with the compose mode so that it’s possible to stream of consciousness your instructions into it?
I think this would be a game-changer in terms of compose mode usability! haha if I could just send you a pull request, I would even do this for free for you. That level of interactivity would be so incredible!!
I don’t think Cursor will add additional external STT models, and using existing API services will directly increase Cursor’s cost. But it sounds like you already have one. If you need one with context there are many options, local or service-based. I can suggest TalkTastic, if you have a setapp account there is superwhisper or Flow, use Use code CYBERMONDAY by Dec 8 to get 33% off.
Hmm but the issue is that I want a STT that I’m comfortable with with the compose UI. I think that could be a game-changer for usability.
It could even be something with usage-based pricing like o1?