jhj0517

Results 115 comments of jhj0517
trafficstars

Hi, it seems like whisper hallucination. - Related discussion - https://github.com/openai/whisper/discussions/679 Many of possible solutions are discussed here. You can try - Set `condition_on_previous_text` to False - Tune `no_speech_threshold` and...

Silero VAD Filter is added in #153. Open the "Advanced Parameters" tab in the WebUI, and check "Enable Silero VAD Filter". If the hallucination still occurs, uncheck "Condition On Previous...

@lgs777 Thanks for pointing this out, I think this is a pretty notable issue. I'll just update the default model to `large-v2` for now.

I just added BGM separation pre processing to reduce such hallucinations in #267. ![image](https://github.com/user-attachments/assets/b8428802-3267-4864-b11f-4f6518b49511) It gave me really better result on my test when the audio includes bgm, please feel...

@mark-wd Thanks for pointing that out. I updated some labels for clearer use of submodels in #308. If anyone has suggestions for better clarification, I'd appreciate it.

Hi! We've attempted to address this in #15. Right now, we're calling `torch.cuda.empty_cache()` after each transcription. If anyone has idea or PR for a better solution, it would be very...

Thanks for sharing your experience! According to [here](https://pytorch.org/docs/stable/notes/hip.html#memory-management), "occupied GPU memory by tensors will not be freed" is normal behavior because it only frees GPU memory cache that can be...

Hi! 1. Here's the [Dockerfile](https://github.com/jhj0517/Whisper-WebUI/blob/master/Dockerfile) you can use. 2. Regarding the REST API, you can build it directly in Gradio. I'd recommend reading this Gradio [documentation](https://www.gradio.app/guides/sharing-your-app#api-page). You can view the...

Hi, According to [gradio #2986](https://github.com/gradio-app/gradio/issues/2986), it seems that some JavaScript is problematic on your browser to run gradio . To fix this, 1. Disable All Chrome Extensions : some extensions...

Hi, sorry for the late response. This issue seems to be the same with #77, if it only happens when you upload a huge file. I guess this is related...