pyannote-audio

Neural building blocks for speaker diarization:

• speech activity detection

• speaker change detection

• overlapped speech detection

• speaker embedding

• speaker diarization pipeline

Installation

# create a conda environment with Python 3.6 or later$ conda create --name pyannote python=3.6
$ source activate pyannote# install pytorch following official instructions from https://pytorch.org/# install from source in the "develop" branch$ git clone https://github.com/pyannote/pyannote-audio.git
$ cd pyannote-audio
$ git checkout develop
$ pip install .

Citation

If you use pyannote.audio please use the following citations.

• Speech activity and speaker change detection

  @inproceedings{Yin2017,  Author = {Ruiqing Yin and Herv'e Bredin and Claude Barras},  Title = {{Speaker Change Detection in Broadcast TV using Bidirectional Long Short-Term Memory Networks}},  Booktitle = {{18th Annual Conference of the International Speech Communication Association, Interspeech 2017}},  Year = {2017},  Month = {August},  Address = {Stockholm, Sweden},  Url = {https://github.com/yinruiqing/change_detection}}

• Speaker embedding

  @inproceedings{Bredin2017,    author = {Herv'{e} Bredin},    title = {{TristouNet: Triplet Loss for Speaker Turn Embedding}},    booktitle = {42nd IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2017},    year = {2017},    url = {http://arxiv.org/abs/1609.04301},
  }

• Speaker diarization pipeline

  @inproceedings{Yin2018,  Author = {Ruiqing Yin and Herv'e Bredin and Claude Barras},  Title = {{Neural Speech Turn Segmentation and Affinity Propagation for Speaker Diarization}},  Booktitle = {{19th Annual Conference of the International Speech Communication Association, Interspeech 2018}},  Year = {2018},  Month = {September},  Address = {Hyderabad, India},
  }

Tutorials

⚠️ These tutorials assumes that you installed the develop branch of pyannote.audio.
⚠️ They are most likely broken in pyannote.audio 1.x.

• Feature extraction

• Models

• Training LSTM-based speech activity detection

• Training LSTM-based speaker change detection

• Training LSTM-based overlapped speech detection

• Training LSTM-based speaker embedding

• Pipelines

• Tuning speech activity detection pipeline

• Tuning speaker diarization pipeline

• In-house datasets

• Applying pre-trained models on your own data

• Training models on your own data

Documentation

Part of the API is described in this tutorial.
Other than that, there is still a lot to do (contribute?) documentation-wise...