CN-Celeb plugin for pyannote.database

This package provides an implementation of the speaker verification protocols used in the CNCeleb paper.

Citation

Please cite the following reference if your research relies on the CNCeleb dataset:

@inproceedings{fan2020cn,
  title={CN-CELEB: a challenging Chinese speaker recognition dataset},
  author={Fan, Yue and Kang, JW and Li, LT and Li, KC and Chen, HL and
          Cheng, ST and Zhang, PY and Zhou, ZY and Cai, YQ and Wang, Dong},
  booktitle={ICASSP 2020-2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)},
  pages={7604--7608},
  year={2020},
  organization={IEEE}
}

Please cite the following references if your research relies on this package. This is where the whole pyannote.database framework was first introduced:

@inproceedings{pyannote.metrics,
  author = {Herv\'e Bredin},
  title = {{pyannote.metrics: a toolkit for reproducible evaluation, diagnostic, and error analysis of speaker diarization systems}},
  booktitle = {{Interspeech 2017, 18th Annual Conference of the International Speech Communication Association}},
  year = {2017},
  month = {August},
  address = {Stockholm, Sweden},
  url = {http://pyannote.github.io/pyannote-metrics},
}

Installation

• Install this package

$ pip install pyannote.db.cnceleb

• Download CN-Celeb dataset to obtain this kind of directory structure:

/path/to/CN-Celeb/CN-Celeb_flac/data/id00001/entertainment-01-001.flac
...
/path/to/CN-Celeb/CN-Celeb2_flac/data/id10000/vlog-01-001.flac
...

• Update ~/.pyannote/database.yml to look like this:

Databases:
  CNCeleb:
    - /path/to/CN-Celeb/CN-Celeb_flac/data/{uri}.flac
    - /path/to/CN-Celeb/CN-Celeb2_flac/data/{uri}.flac

Usage

from pyannote.database import get_protocol
protocol = get_protocol('CNCeleb.SpeakerVerification.CNCeleb')

One can use protocol.train generator to train the background model.

for training_file in protocol.train():
    uri = training_file['uri']
    print('Current filename is {0}.'.format(uri))

    # "who speaks when" as a pyannote.core.Annotation instance
    annotation = training_file['annotation']
    for segment, _, speaker in annotation.itertracks(yield_label=True):
        print('{0} speaks between t={1:.1f}s and t={2:.1f}s.'.format(
            speaker, segment.start, segment.end))
   
    break  # this should obviously be replaced
           # by the actual background training