Distinguishing Neural Speech Synthesis Models Through Fingerprints in Speech Waveforms

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Zhang, Chu Yuan, Yi, Jiangyan, Tao, Jianhua, Wang, Chenglong, Yan, Xinrui
Format: Preprint
Veröffentlicht: 2023
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866914834733334528
author Zhang, Chu Yuan
Yi, Jiangyan
Tao, Jianhua
Wang, Chenglong
Yan, Xinrui
author_facet Zhang, Chu Yuan
Yi, Jiangyan
Tao, Jianhua
Wang, Chenglong
Yan, Xinrui
contents Recent strides in neural speech synthesis technologies, while enjoying widespread applications, have nonetheless introduced a series of challenges, spurring interest in the defence against the threat of misuse and abuse. Notably, source attribution of synthesized speech has value in forensics and intellectual property protection, but prior work in this area has certain limitations in scope. To address the gaps, we present our findings concerning the identification of the sources of synthesized speech in this paper. We investigate the existence of speech synthesis model fingerprints in the generated speech waveforms, with a focus on the acoustic model and the vocoder, and study the influence of each component on the fingerprint in the overall speech waveforms. Our research, conducted using the multi-speaker LibriTTS dataset, demonstrates two key insights: (1) vocoders and acoustic models impart distinct, model-specific fingerprints on the waveforms they generate, and (2) vocoder fingerprints are the more dominant of the two, and may mask the fingerprints from the acoustic model. These findings strongly suggest the existence of model-specific fingerprints for both the acoustic model and the vocoder, highlighting their potential utility in source identification applications.
format Preprint
id arxiv_https___arxiv_org_abs_2309_06780
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Distinguishing Neural Speech Synthesis Models Through Fingerprints in Speech Waveforms
Zhang, Chu Yuan
Yi, Jiangyan
Tao, Jianhua
Wang, Chenglong
Yan, Xinrui
Sound
Audio and Speech Processing
Recent strides in neural speech synthesis technologies, while enjoying widespread applications, have nonetheless introduced a series of challenges, spurring interest in the defence against the threat of misuse and abuse. Notably, source attribution of synthesized speech has value in forensics and intellectual property protection, but prior work in this area has certain limitations in scope. To address the gaps, we present our findings concerning the identification of the sources of synthesized speech in this paper. We investigate the existence of speech synthesis model fingerprints in the generated speech waveforms, with a focus on the acoustic model and the vocoder, and study the influence of each component on the fingerprint in the overall speech waveforms. Our research, conducted using the multi-speaker LibriTTS dataset, demonstrates two key insights: (1) vocoders and acoustic models impart distinct, model-specific fingerprints on the waveforms they generate, and (2) vocoder fingerprints are the more dominant of the two, and may mask the fingerprints from the acoustic model. These findings strongly suggest the existence of model-specific fingerprints for both the acoustic model and the vocoder, highlighting their potential utility in source identification applications.
title Distinguishing Neural Speech Synthesis Models Through Fingerprints in Speech Waveforms
topic Sound
Audio and Speech Processing
url https://arxiv.org/abs/2309.06780