Voice Privacy - leveraging multi-scale blocks with ECAPA-TDNN SE-Res2NeXt extension for speaker anonymization

Khamsehashari, Razieh; Sinha, Yamini; Hintz, Jan; Ghosh, Suhita; Polzehl, Tim; Franzreb, Clarlos; Stober, Sebastian; Siegert, Ingo

doi:10.21437/spsc.2022-8

Cited by 3 publications

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some however, such as [7], propose alternatives to the bottleneck features. For the speaker embedding, [8] proposes switching to ECAPA and [9] reports increased speaker representation capabilities when both ECAPA and x-vectors are used together.…”

Section: Related Work 21 Vpc Baseline B1 and Its Derivativesmentioning

confidence: 99%

Evaluation of the Speech Resynthesis Capabilities of the VoicePrivacy Baseline B1

Gaznepoglu,

Peters

2023

3rd Symposium on Security and Privacy in Speech Communication

View full text Add to dashboard Cite

Speaker anonymization systems continue to improve their ability to obfuscate the original speaker characteristics in a speech signal, but often create processing artifacts and unnatural sounding voices as a tradeoff. Many of those systems stem from the VoicePrivacy Challenge (VPC) Baseline B1, using a neural vocoder to synthesize speech from an F0, x-vectors and bottleneck features-based speech representation. Inspired by this, we investigate the reproduction capabilities of the aforementioned baseline, to assess how successful the shared methodology is in synthesizing human-like speech. We use four objective metrics to measure speech quality, waveform similarity, and F0 similarity. Our findings indicate that both the speech representation and the vocoder introduces artifacts, causing an unnatural perception. A MUSHRA-like listening test on 18 subjects corroborate our findings, motivating further research on the analysis and synthesis components of the VPC Baseline B1.

show abstract

Section: Related Work 21 Vpc Baseline B1 and Its Derivativesmentioning

confidence: 99%

Evaluation of the Speech Resynthesis Capabilities of the VoicePrivacy Baseline B1

Gaznepoglu,

Peters

2023

3rd Symposium on Security and Privacy in Speech Communication

View full text Add to dashboard Cite

show abstract

Study of ECAPA-TDNN Models for Spoken Language Identification Task

Mandal,

Mukherjee

2023

2023 IEEE World Conference on Applied Intelligence and Computing (AIC)

View full text Add to dashboard Cite

Speaker Anonymization Using Orthogonal Householder Neural Network

Miao,

Wang,

Cooper

et al. 2023

IEEE/ACM Trans. Audio Speech Lang. Process.

View full text Add to dashboard Cite

Speaker anonymization aims to conceal a speaker's identity while preserving content information in speech. Current mainstream neural-network speaker anonymization systems disentangle speech into prosody-related, content, and speaker representations. The speaker representation is then anonymized by a selection-based speaker anonymizer that uses a mean vector over a set of randomly selected speaker vectors from an external pool of English speakers. However, the resulting anonymized vectors are subject to severe privacy leakage against powerful attackers, reduction in speaker diversity, and language mismatch problems for unseen-language speaker anonymization. To generate diverse, language-neutral speaker vectors, this paper proposes an anonymizer based on an orthogonal Householder neural network (OHNN). Specifically, the OHNN acts like a rotation to transform the original speaker vectors into anonymized speaker vectors, which are constrained to follow the distribution over the original speaker vector space. A basic classification loss is introduced to ensure that anonymized speaker vectors from different speakers have unique speaker identities. To further protect speaker identities, an improved classification loss and similarity loss are used to push original-anonymized sample pairs away from each other. Experiments on VoicePrivacy Challenge datasets in English and the AISHELL-3 dataset in Mandarin demonstrate the proposed anonymizer's effectiveness.

show abstract

Voice Privacy - leveraging multi-scale blocks with ECAPA-TDNN SE-Res2NeXt extension for speaker anonymization

Cited by 3 publications

References 0 publications

Evaluation of the Speech Resynthesis Capabilities of the VoicePrivacy Baseline B1

Evaluation of the Speech Resynthesis Capabilities of the VoicePrivacy Baseline B1

Study of ECAPA-TDNN Models for Spoken Language Identification Task

Speaker Anonymization Using Orthogonal Householder Neural Network

Contact Info

Product

Resources

About