[HTML][HTML] Speaker anonymization by modifying fundamental frequency and x-vector singular value

CO Mawalim, K Galajit, J Karnjana, S Kidani… - Computer Speech & …, 2022 - Elsevier
Speaker anonymization is a method of protecting voice privacy by concealing individual
speaker characteristics while preserving linguistic information. The VoicePrivacy Challenge …

Improving personalized predictions through nonlinear individual-specific parameters in neural network-based mixed effect models

B Ramesh, D Dhanya, LR Bindu, G Vikram… - AIP Conference …, 2024 - pubs.aip.org
For sequential tasks, such as daily mood ratings, personalized prediction—a machine
learning method—can be useful since it predicts future observations based on a person's …

Issues of bilingualism in likelihood ratio-based forensic voice comparison

JH Lo - 2021 - etheses.whiterose.ac.uk
Situated at the intersection of forensic speech science and bilingualism, this thesis focuses
on the issues of language and language mismatch in forensic voice comparison (FVC) and …

[PDF][PDF] Impact of the mismatches in long-term acoustic features upon different-speaker ASR scores

C Xu, P Foulkes, P Harrison, V Hughes… - 31st IAFPA …, 2023 - iafpa2023.uzh.ch
Automatic speaker recognition (ASR) systems usually take a pair of speech recordings as
input, extract their speaker embeddings using deep learning (eg x-vectors; Snyder et al …

Data driven and signal processing techniques for audio forensics

C Borrelli - 2021 - politesi.polimi.it
The recent developments and diffusion of audio recording devices, audio editing tools and
speech synthesis techniques have opened questions about how to verify the authenticity …