Replies: 1 comment 1 reply
-
The model would need to be implemented in C++. We'd happily welcome contributions doing that, if anyone wants to take up the cause of porting this from Torch to GGML. |
Beta Was this translation helpful? Give feedback.
1 reply
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
https://github.com/FunAudioLLM/SenseVoice for reference. It claims to have much better latency for the small model, and better performance on non-english languages, especially chinese/cantonese.
Beta Was this translation helpful? Give feedback.
All reactions