Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | https://arxiv.org/abs/2409.17285 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866910911393955840 |
|---|---|
| author | Jung, Jee-weon Wu, Yihan Wang, Xin Kim, Ji-Hoon Maiti, Soumi Matsunaga, Yuta Shim, Hye-jin Tian, Jinchuan Evans, Nicholas Chung, Joon Son Zhang, Wangyou Um, Seyun Takamichi, Shinnosuke Watanabe, Shinji |
| author_facet | Jung, Jee-weon Wu, Yihan Wang, Xin Kim, Ji-Hoon Maiti, Soumi Matsunaga, Yuta Shim, Hye-jin Tian, Jinchuan Evans, Nicholas Chung, Joon Son Zhang, Wangyou Um, Seyun Takamichi, Shinnosuke Watanabe, Shinji |
| contents | This paper introduces SpoofCeleb, a dataset designed for Speech Deepfake Detection (SDD) and Spoofing-robust Automatic Speaker Verification (SASV), utilizing source data from real-world conditions and spoofing attacks generated by Text-To-Speech (TTS) systems also trained on the same real-world data. Robust recognition systems require speech data recorded in varied acoustic environments with different levels of noise to be trained. However, current datasets typically include clean, high-quality recordings (bona fide data) due to the requirements for TTS training; studio-quality or well-recorded read speech is typically necessary to train TTS models. Current SDD datasets also have limited usefulness for training SASV models due to insufficient speaker diversity. SpoofCeleb leverages a fully automated pipeline we developed that processes the VoxCeleb1 dataset, transforming it into a suitable form for TTS training. We subsequently train 23 contemporary TTS systems. SpoofCeleb comprises over 2.5 million utterances from 1,251 unique speakers, collected under natural, real-world conditions. The dataset includes carefully partitioned training, validation, and evaluation sets with well-controlled experimental protocols. We present the baseline results for both SDD and SASV tasks. All data, protocols, and baselines are publicly available at https://jungjee.github.io/spoofceleb. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2409_17285 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | SpoofCeleb: Speech Deepfake Detection and SASV In The Wild Jung, Jee-weon Wu, Yihan Wang, Xin Kim, Ji-Hoon Maiti, Soumi Matsunaga, Yuta Shim, Hye-jin Tian, Jinchuan Evans, Nicholas Chung, Joon Son Zhang, Wangyou Um, Seyun Takamichi, Shinnosuke Watanabe, Shinji Sound Artificial Intelligence Audio and Speech Processing This paper introduces SpoofCeleb, a dataset designed for Speech Deepfake Detection (SDD) and Spoofing-robust Automatic Speaker Verification (SASV), utilizing source data from real-world conditions and spoofing attacks generated by Text-To-Speech (TTS) systems also trained on the same real-world data. Robust recognition systems require speech data recorded in varied acoustic environments with different levels of noise to be trained. However, current datasets typically include clean, high-quality recordings (bona fide data) due to the requirements for TTS training; studio-quality or well-recorded read speech is typically necessary to train TTS models. Current SDD datasets also have limited usefulness for training SASV models due to insufficient speaker diversity. SpoofCeleb leverages a fully automated pipeline we developed that processes the VoxCeleb1 dataset, transforming it into a suitable form for TTS training. We subsequently train 23 contemporary TTS systems. SpoofCeleb comprises over 2.5 million utterances from 1,251 unique speakers, collected under natural, real-world conditions. The dataset includes carefully partitioned training, validation, and evaluation sets with well-controlled experimental protocols. We present the baseline results for both SDD and SASV tasks. All data, protocols, and baselines are publicly available at https://jungjee.github.io/spoofceleb. |
| title | SpoofCeleb: Speech Deepfake Detection and SASV In The Wild |
| topic | Sound Artificial Intelligence Audio and Speech Processing |
| url | https://arxiv.org/abs/2409.17285 |