You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Could FluidVoice support custom Whisper models for dictation?
#1279
Hi, thanks for FluidVoice! I dictate in Swiss German, which works best with a fine-tuned model like Flix-AI/flix-swissgerman-full (Whisper large-v3).
As far as I can tell, 1.6.12 can only use the built-in Whisper models, since it checks each file's name and size. Would you be open to letting users pick a local .gguf file as a voice model? It would also help anyone who uses a fine-tuned model for another dialect or a specialised vocabulary.
I haven't tried converting this model to GGUF yet, but I'm happy to test that if the idea sounds reasonable. Is there a better way to do this that I've missed?
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
Uh oh!
There was an error while loading. Please reload this page.
Hi, thanks for FluidVoice! I dictate in Swiss German, which works best with a fine-tuned model like Flix-AI/flix-swissgerman-full (Whisper large-v3).
As far as I can tell, 1.6.12 can only use the built-in Whisper models, since it checks each file's name and size. Would you be open to letting users pick a local
.gguffile as a voice model? It would also help anyone who uses a fine-tuned model for another dialect or a specialised vocabulary.I haven't tried converting this model to GGUF yet, but I'm happy to test that if the idea sounds reasonable. Is there a better way to do this that I've missed?
All reactions