3 ms·
Ive been using the text-to-speech model Whisper, from fdriod. Its rather small and all processing is done locally on my phone. Its pretty good.
by username135 1y ago
Ive been using the text-to-speech model Whisper, from fdriod. Its rather small and all processing is done locally on my phone. Its pretty good.
- sebastiennight 1y agoYou mean speech-to-text, right? For dictation/transcription? It is pretty good indeed (despite the ~30sec input limit), but this feels unrelated to the topic at hand.