This is a demonstration of using Whisper to recognize voice commands in the browser.
Usage:
- Hold the button and say a chess move (e.g. "e2-e4")
- Release the button and wait for the move to be recognized
- Repeat
- "d4"
- "e2 e4"
- "Knight to f3"
- "Bishop to b5"
currently do not work, but can be easily implemented. The main purpose of this example
is to demonstrate the capabilities of Whisper and it's application in the browser.
Features:
- Model quantization for reduced memory footprint (~42MB)
- Grammar-based sampling for improved recognition accuracy
More examples: main | bench | stream | command | talk |
Whisper model:
Status: not started
[The moves will be displayed here]
Debug output:
Troubleshooting
The page does some heavy computations, so make sure:
- To use a modern web browser (e.g. Chrome, Firefox)
- Your browser supports WASM Fixed-width SIMD
|
Build time: be5cd736 |
Commit subject: |
Source Code |
|
Commit hash: