yat011 / pyannote-audio

Neural building blocks for speaker diarization: speech activity detection, speaker change detection, speaker embedding

Announcement

Open Phd/postdoc positions at LIMSI combining machine learning, NLP, speech processing, and computer vision.

pyannote-audio

Audio processing

Installation

$ conda create --name pyannote python=3.5 anaconda
$ source activate pyannote
$ conda install -c conda-forge yaafe
$ pip install -U pip setuptools
$ pip install pyannote.audio

Citation

If you use pyannote.audio in your research, please use the following citation (until a regular paper is published):

@misc{pyannote.audio,
  author = {Bredin, H.},
  title = {pyannote.audio},
  year = {2017},
  publisher = {GitHub},
  journal = {GitHub repository},
  howpublished = {\url{https://github.com/pyannote/pyannote-audio}},
}

Tutorials

• Feature extraction
• LSTM-based speech activity detection
• LSTM-based speaker change detection
• TristouNet neural speech turn embedding

Documentation

The API is unfortunately not documented yet.