Can Large-Scale Vocoded Spoofed Data Improve Speech Spoofing Countermeasure with a Self-Supervised Front End?

DOI DOI PDF 参考文献28件 オープンアクセス

書誌事項

公開日
2024-04-14
資源種別
journal article
権利情報
  • https://doi.org/10.15223/policy-029
  • https://doi.org/10.15223/policy-037
DOI
  • 10.1109/icassp48485.2024.10446331
  • 10.48550/arxiv.2309.06014
公開者
IEEE

説明

A speech spoofing countermeasure (CM) that discriminates between unseen spoofed and bona fide data requires diverse training data. While many datasets use spoofed data generated by speech synthesis systems, it was recently found that data vocoded by neural vocoders were also effective as the spoofed training data. Since many neural vocoders are fast in building and generation, this study used multiple neural vocoders and created more than 9,000 hours of vocoded data on the basis of the VoxCeleb2 corpus. This study investigates how this large-scale vocoded data can improve spoofing countermeasures that use data-hungry self-supervised learning (SSL) models. Experiments demonstrated that the overall CM performance on multiple test sets improved when using features extracted by an SSL model continually trained on the vocoded data. Further improvement was observed when using a new SSL distilled from the two SSLs before and after the continual training. The CM with the distilled SSL outperformed the previous best model on challenging unseen test sets, including the ASVspoof 2019 logical access, WaveFake, and In-the-Wild.

To appear in ICASSP 2024. code on github: https://github.com/nii-yamagishilab/project-NN-Pytorch-scripts/tree/master/project/10-asvspoof-vocoded-trn-ssl

収録刊行物

参考文献 (28)*注記

もっと見る

関連プロジェクト

もっと見る

キーワード

詳細情報 詳細情報について

問題の指摘

ページトップへ