whisper-web
whisper-web copied to clipboard
xenova
ML-powered speech recognition directly in your browser
If i change the transformers to a new version i get that if i revert back to the "@huggingface/transformers": "^3.0.0-alpha.19" and do rm -rf node_modules rm package-lock.json npm cache clean...
Hello How are you? Thanks for contributing to this project. Is it possible to run an Whisper-tiny model or Whisper-base model on a mobile phone web directly?
i was struggling with the audio quality from my laptop microphone so I added the option to choose the device 👾
Hi @xenova, As promised, I'll try to contribute some of the changes from my fork back to the main project. I suggest to start from your branch `experimental-webgpu` (which seems...
Should be merged after #57 and #58 This small PR fixes all the errors and warnings raised by `npm run lint`. I also fixed some formatting issues with `npm run...
This PR should be merged after #57 It removes the requirement to have WebGPU support in the browser and adds a checkbox to choose between GPU or no GPU. This...
Hello, You should embed your application into an ElectronJS shell. This would demo a fully locally Speech2Text with the file management into the Chtomium. It would lead to a clean,...
When running the WebGPU demo at https://huggingface.co/spaces/Xenova/whisper-webgpu, the model fails to load because of an error:  Apparently, the method `GPUAdapter.requestAdapterInfo` on adapter is deprecated and removed in favor of...
