This repository contains the framework for training speaker verification model described in [2]
with score normalization post-processing described in [3].

pip install -r requirements.txt

1. Generate train, validate list (if ratio == -1, take 3 files for each speaker for validate)

python src/dataprep.py --generate --split_ratio -1 --config yaml/configuration.yaml

In addition to the Python dependencies, wget and ffmpeg must be installed on the system.

Set cuda usage

export CUDA_VISIBLE_DEVICES=5

then add the device="cuda:5" to args
Single GPU

CUDA_VISIBLE_DEVICES=0 python main.py --do_train --config yaml/configuration.yaml

Data parallel

CUDA_VISIBLE_DEVICES=0,1,2 python main.py --do_train --config yaml/configuration.yaml --data_parallel

Distributed

CUDA_VISIBLE_DEVICES=0,1,2 python main.py --do_train --config yaml/configuration.yaml --distributed --mixedprec --distributed_backend nccl --port 10001

Note: the best model is automatically saved during the training process, if the initial_model is not provided, automatically load from the best_state weight if possible. add --augment to train with augment data

1. prepare cohorts

CUDA_VISIBLE_DEVICES=0 python main.py --do_infer --prepare --config yaml/configuration.yaml

2. Evaluate and tune thresholds

CUDA_VISIBLE_DEVICES=0 python main.py --do_infer --eval --config yaml/configuration.yaml

3. Run on Test set

CUDA_VISIBLE_DEVICES=0 python main.py --do_infer --test --config yaml/configuration.yaml

[1] In defence of metric learning for speaker recognition

@inproceedings{chung2020in,
    title={In defence of metric learning for speaker recognition},
    author={Chung, Joon Son and Huh, Jaesung and Mun, Seongkyu and Lee, Minjae and Heo, Hee Soo and Choe, Soyeon and Ham, Chiheon and Jung, Sunghwan and Lee, Bong-Jin and Han, Icksang},
    booktitle={Interspeech},
    year={2020}
}

[2] Clova baseline system for the VoxCeleb Speaker Recognition Challenge 2020

@article{heo2020clova,
    title={Clova baseline system for the {VoxCeleb} Speaker Recognition Challenge 2020},
    author={Heo, Hee Soo and Lee, Bong-Jin and Huh, Jaesung and Chung, Joon Son},
    journal={arXiv preprint arXiv:2009.14153},
    year={2020}
}