Whispering

Streaming transcriber with whisper. Enough machine power is needed to transcribe in real time.

Setup

pip install -U git+https://github.com/shirayu/whispering.git@v0.5.1

# If you use GPU, install proper torch and torchaudio
# Check https://pytorch.org/get-started/locally/
# Example : torch for CUDA 11.6
pip install -U torch torchaudio --extra-index-url https://download.pytorch.org/whl/cu116

If you get OSError: PortAudio library not found in Linux, install "PortAudio".

sudo apt -y install portaudio19-dev

Example of microphone

# Run in English
whispering --language en --model tiny

• --help shows full options
• --model set the model name to use. Larger models will be more accurate, but may not be able to transcribe in real time.
• --language sets the language to transcribe. The list of languages are shown with whispering -h
• --no-progress disables the progress message
• -t sets temperatures to decode. You can set several like -t 0.0 -t 0.1 -t 0.5, but too many temperatures exhaust decoding time
• --debug outputs logs for debug
• --no-vad disables VAD (Voice Activity Detection). This forces whisper to analyze non-voice activity sound period
• --output sets output file (Default: Standard output)

Parse interval

Without --allow-padding, whispering just performs VAD for the period, and when it is predicted as "silence", it will not be passed to whisper. If you want to change the VAD interval, change -n.

If you want quick response, set small -n and add --allow-padding. However, this may sacrifice the accuracy.

whispering --language en --model tiny -n 20 --allow-padding

Example of web socket

⚠ No security mechanism. Please make secure with your responsibility.

Run with --host and --port.

Host

whispering --language en --model tiny --host 0.0.0.0 --port 8000

Client

whispering --host ADDRESS_OF_HOST --port 8000 --mode client

You can set -n, --allow-padding and other options.

License

• MIT License
• Some codes are ported from the original whisper. Its license is also MIT License