Low-latency speaker spotting with online diarization and detection
$ conda create --name pyannote python=3.6 anaconda
$ source activate pyannote
$ conda install -c yaafe yaafe=0.65
$ pip install pyannote.audio
$ pip install pyannote.db.odessa.ami
$ git clone
If you use this tool, please cite the following paper:
@inproceedings{patino2018low,
title={{Low-latency Speaker Spotting with Online Diarization and Detection}},
author={Patino, Jose and Yin, Ruiqing and Delgado, H\'ector and Bredin, Herv\'e and Komaty, Alain and Wisniewski, Guillaume and Barras, Claude and Evans, Nicholas and Marcel, S\'ebastien },
booktitle={The Speaker and Language Recognition Workshop (Odyssey 2018)},
year={2018}
}
Please follow the documentation
in pyannote.audio
to extract embeddings. A pretrained model (trained and validated in AMI database) is available in
$ export $EMBEDDING_DIR=/path/of/extracted/embeddings
$ conda create --name pyannote python=3.6 anaconda
$ export $EMBEDDING_DIR=/path/of/extracted/embeddings
$ conda create --name pyannote python=3.6 anaconda