SpoofCeleb: Speech Deepfake Detection and SASV In The Wild

TL;DR

SpoofCeleb is a large, real-world dataset for speech deepfake detection and speaker verification robustness, created by transforming VoxCeleb1 data and training multiple TTS systems, enabling more realistic and diverse evaluation.

Contribution

The paper introduces SpoofCeleb, a comprehensive dataset for SDD and SASV, generated from real-world data and multiple TTS systems, addressing limitations of existing datasets.

Findings

Over 2.5 million utterances from 1,251 speakers.

Baseline results provided for SDD and SASV tasks.

Dataset and protocols are publicly available.

Abstract

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.