Descriptor:: Extended-Length Audio Dataset for Synthetic Voice Detection and Speaker Recognition (ELAD-SVDSR)

TL;DR

ELAD-SVDSR introduces an extended-duration audio dataset for synthetic voice detection and speaker recognition, addressing the need for long-context data and synthetic samples. The corpus comprises 45-minute recordings from 36 participants collected under five microphone conditions, plus 20 deepfake voices generated with a TTS system to enable robust benchmarking. Ethical safeguards include IRB approval (No. 24-42), informed consent, anonymization, and secure data handling. VeriSpeak-based evaluations show meaningful synthetic-original similarity (mean around 37.2%), and comparisons to published datasets indicate the dataset provides challenging deepfakes for benchmarking detection methods, with implications for audioForensics and biometric security.

Abstract

This paper introduces the Extended Length Audio Dataset for Synthetic Voice Detection and Speaker Recognition (ELAD SVDSR), a resource specifically designed to facilitate the creation of high quality deepfakes and support the development of detection systems trained against them. The dataset comprises 45 minute audio recordings from 36 participants, each reading various newspaper articles recorded under controlled conditions and captured via five microphones of differing quality. By focusing on extended duration audio, ELAD SVDSR captures a richer range of speech attributes such as pitch contours, intonation patterns, and nuanced delivery enabling models to generate more realistic and coherent synthetic voices. In turn, this approach allows for the creation of robust deepfakes that can serve as challenging examples in datasets used to train and evaluate synthetic voice detection methods. As part of this effort, 20 deepfake voices have already been created and added to the dataset to showcase its potential. Anonymized metadata accompanies the dataset on speaker demographics. ELAD SVDSR is expected to spur significant advancements in audio forensics, biometric security, and voice authentication systems.

Figures (3)

• Figure 1: Live data collection Process
• Figure 2: Microphones and recording devices used in this study (from top to bottom): Audio-Technica AT2020, Shure SM58, TOZO A1 earbuds, Inni Oasis R1 recorder, and the ZIPCIDE Spy Pen.
• Figure 3: Directory Architecture for ELAD-SVDSR