Bug 251939
| Summary: | [Meta][GTK][WPE] Implement SpeechRecognition API | ||
|---|---|---|---|
| Product: | WebKit | Reporter: | ChangSeok Oh <changseok> |
| Component: | WebKitGTK | Assignee: | ChangSeok Oh <changseok> |
| Status: | NEW | ||
| Severity: | Normal | CC: | bugs-noreply, philn |
| Priority: | P2 | ||
| Version: | WebKit Nightly Build | ||
| Hardware: | Unspecified | ||
| OS: | Unspecified | ||
| Bug Depends on: | 260709 | ||
| Bug Blocks: | |||
ChangSeok Oh
I will explore and examine possibilities for bringing SpeechRecognition to the gtk/wpe port in this bug.
| Attachments | ||
|---|---|---|
| Add attachment proposed patch, testcase, etc. |
ChangSeok Oh
My working branch is https://github.com/shivamidow/WebKit/tree/eng/prototype-speech-to-text-with-whisper.
It is close to being ready for review. For those who are interested, I leave build instructions.
Once you add my branch to your repo,
$ git submodule update --init
$ cd Source/ThirdParty/whisper.cpp
$ make base.en # download language models, English only now.
$ cd ../../..
$ Tools/Script/build-webkit --gtk
Once webkit is built,
$ WEBKIT_WHISPER_MODEL_PATH=Source/ThirdParty/whisper.cpp/models Tools/Scripts/run-minibrowser --gtk https://www.google.com/intl/en/chrome/demos/speech.html
And enjoy!