[FEAT] How about adding a dictation using the OpenAI's text-to-spech?
mykolasolodukha opened this issue · comments
Mykola Solodukha commented
Docs: https://platform.openai.com/docs/guides/text-to-speech
Paired with the existing feature called "sound on error" (or something, I don't recall the exact name) it would give quite an immersive experience, so you can e.g. close your eyes and think about the keyboard at hand (quite literally), rather than being forced to read first, then type.
Rafael Aguilar commented
for a quick poc we could use we speech api,
https://www.assemblyai.com/blog/javascript-text-to-speech-easy-way/
may be that's enough for some people, beses no t worrying about latency