Wake Word Tagger

The Wake Word Tagger is a web application that allows you to tag and classify audio samples as "Wake Word", "NOT Wake Word", or other categories.

It also allows you to specify the speaker's gender and the type of silence in the audio sample.

Features

• Browse and play audio samples (.wav files) in a web interface
• Tag each audio sample with a classification (Wake Word, NOT Wake Word, or Unknown)
• Specify the speaker's gender (Male, Female, or Unknown)
• Specify the type of silence in the audio sample (Music/TV, Noise, Human Non-Speech, Silence, or Unknown)
• Navigate through the audio samples using "Previous" and "Next" buttons
• Automatically save the tags and metadata to a JSON database file

Usage

git clone https://github.com/your-username/wake-word-tagger

cd wake-word-tagger

pip install -r requirements.txt

python app.py --folder /path/to/audio/files --db /path/to/tags.json

Replace /path/to/audio/files with the actual path to the folder containing the .wav files, and tags.json with the JSON database file.

If not specified will default to ~/.local/share/mycroft/listener/wake_words (default save path for ovos-listener)

Open your web browser and navigate to http://localhost:5000 to access the Wake Word Tagger interface.

Contributing

If you find any issues or have suggestions for improvements, feel free to open an issue or submit a pull request.