Distinguishing Neural Speech Synthesis Models Through Fingerprints in Speech Waveforms

TL;DR

This paper investigates the presence of unique, model-specific fingerprints in speech waveforms generated by neural speech synthesis models, revealing that vocoders and acoustic models leave distinguishable traces useful for source attribution.

Contribution

It demonstrates the existence of distinct fingerprints from both acoustic models and vocoders in synthesized speech, with vocoder fingerprints being more dominant, advancing source identification methods.

Findings

Vocoder and acoustic model fingerprints are present in speech waveforms.

Vocoder fingerprints are more prominent and can mask acoustic model fingerprints.

Findings support the use of model fingerprints for source attribution in synthesized speech.

Abstract

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.