Realtime Speech-to-Text Transcription Demo

A simple realtime speech-to-text transcription from your microhpone using Whisper.

This uses the idea from the whisper_real_time to record audio in a background thread and concatenating the raw bytes over multiple recordings.

Also, my fork of whisper whisper-lang-selection is used to support a selection of languages to detect in your speech.

Setup

Create a virtual environment using the following command:

python -m venv venv

And activate the virtual environment:

source venv/bin/activate # for linux
./venv/Scripts/Activate  # for Windows

Finally, install the required packages using:

pip install -r requirements.txt

Note that you will also need to install ffmpeg to be installed on your system.

Demo

You can run the realtime demo using the following command:

python transcribe.py