Privacy-preserving voice analysis via disentangled representations

R Aloufi, H Haddadi, D Boyle - Proceedings of the 2020 ACM SIGSAC …, 2020‏ - dl.acm.org
Voice User Interfaces (VUIs) are increasingly popular and built into smartphones, home
assistants, and Internet of Things (IoT) devices. Despite offering an always-on convenient …

A study of bias mitigation strategies for speaker recognition

R Peri, K Somandepalli, S Narayanan - Computer Speech & Language, 2023‏ - Elsevier
Speaker recognition is increasingly used in several everyday applications including smart
speakers, customer care centers and other speech-driven analytics. It is crucial to accurately …

Contrastive self-supervised speaker embedding with sequential disentanglement

Y Tu, MW Mak, JT Chien - IEEE/ACM Transactions on Audio …, 2024‏ - ieeexplore.ieee.org
Contrastive self-supervised learning has been widely used in speaker embedding to
address the labeling challenge. Contrastive speaker embedding assumes that the contrast …

Learning disentangled phone and speaker representations in a semi-supervised VQ-VAE paradigm

J Williams, Y Zhao, E Cooper… - ICASSP 2021-2021 …, 2021‏ - ieeexplore.ieee.org
We present a new approach to disentangle speaker voice and phone content by introducing
new components to the VQ-VAE architecture for speech synthesis. The original VQ-VAE …

Contrastive speaker embedding with sequential disentanglement

Y Tu, MW Mak, JT Chien - ICASSP 2024-2024 IEEE …, 2024‏ - ieeexplore.ieee.org
Contrastive speaker embedding assumes that the contrast between the positive and
negative pairs of speech segments is attributed to speaker identity only. However, this …

Random cycle loss and its application to voice conversion

H Sun, D Wang, L Li, C Chen… - IEEE Transactions on …, 2023‏ - ieeexplore.ieee.org
Speech disentanglement aims to decompose independent causal factors of speech signals
into separate codes. Perfect disentanglement benefits to a broad range of speech …

Paralinguistic privacy protection at the edge

R Aloufi, H Haddadi, D Boyle - ACM Transactions on Privacy and …, 2023‏ - dl.acm.org
Voice user interfaces and digital assistants are rapidly entering our lives and becoming
singular touch points spanning our devices. These always-on services capture and transmit …

Exploring disentanglement with multilingual and monolingual VQ-VAE

J Williams, J Fong, E Cooper, J Yamagishi - arxiv preprint arxiv …, 2021‏ - arxiv.org
This work examines the content and usefulness of disentangled phone and speaker
representations from two separately trained VQ-VAE systems: one trained on multilingual …

Large-Scale Functional Connectome Fingerprinting for Generalization and Transfer Learning in Neuroimaging

M Ogg, L Kitchell - bioRxiv, 2024‏ - biorxiv.org
Functional MRI currently supports a limited application space stemming from modest dataset
sizes, large interindividual variability and heterogeneity among scanning protocols. These …