/sed-yamnet-raspberrypi

Sound Event Detection with YAMNet+tw on Raspberry Pi

Primary LanguageJupyter Notebook

Sound Event Detection with YAMNet+tw on Raspberry Pi

Installation

  1. Raspberry Pi 3 Model B+ and Voice HAT microphone are good enough
  2. Headless boot up and SSH in
  3. Clone this repo
    git clone https://github.com/x1001000/sed-yamnet-raspberrypi
  4. Install dependencies, drivers, and configs
    bash sed-yamnet-raspberrypi/install-deps.sh
  5. Reboot to take effect
    sudo reboot
  6. Run SED
    cd && cd sed-yamnet-raspberrypi/yamnet && python3 sed.py
  7. There you go!

References

  1. https://www.tensorflow.org/guide/keras/transfer_learning
  2. https://www.tensorflow.org/tutorials/images/transfer_learning
  3. https://www.tensorflow.org/hub/tutorials/yamnet
  4. https://github.com/tensorflow/models/tree/master/research/audioset
  5. https://github.com/Qengineering/Tensorflow-Raspberry-Pi/
  6. tensorflow/tensorflow#44467
  7. https://github.com/hellmanj/AIY-voice-kit-python/blob/master/HACKING.md
  8. https://people.csail.mit.edu/hubert/pyaudio/