CONFERENCE PROCEEDINGS
Advanced Search
Home > Proceedings > Volume 12506 > Article
Open Access Paper
28 December 2022 Attention enhanced dynamic kernel convolution for TDNN-based speaker verification
Xiaofan Lang, Ya Li
Author Affiliations +
Proceedings Volume 12506, Third International Conference on Computer Science and Communication Technology (ICCSCT 2022); 1250605 (2022) https://doi.org/10.1117/12.2662523
Event: International Conference on Computer Science and Communication Technology (ICCSCT 2022), 2022, Beijing, China
Abstract
Speaker embedding is a state-of-the-art front-end module, which is used to extract discriminative speaker features for speaker-related tasks. The Time Delay Neural Network (TDNN) has been a classical network architecture since it was first applied on speaker related tasks known as X-vector. In this paper, we propose new network structures based on current popular ECAPA-TDNN. We propose a dynamic kernel convolution module to extract features from short-term and long-term context adaptively, thus achieving multi-scale receptive fields. We also apply three enhanced attention modules instead of plain Squeeze-Excitation (SE) layer to realize more efficient information interaction between channels and spaces. The proposed architectures are superior to the most advanced network, with an optimal Equal Error Rate (EER) of 6.40% and a parameters reduction of 6.32%, they also achieve better performances when speaker utterances are shortened.
© (2022) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Citation Download Citation
Xiaofan Lang and Ya Li "Attention enhanced dynamic kernel convolution for TDNN-based speaker verification", Proc. SPIE 12506, Third International Conference on Computer Science and Communication Technology (ICCSCT 2022), 1250605 (28 December 2022); https://doi.org/10.1117/12.2662523
PROCEEDINGS
 7 PAGES
DOWNLOAD PAPER SAVE TO MY LIBRARY
GET CITATION
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Convolution
Speaker recognition
Feature extraction
Performance modeling
Systems modeling
Astatine
Network architectures
RELATED CONTENT
Series arc fault detection of complex branch based on CNN_LSTM...
Proceedings of SPIE (April 25 2022)
The short term network traffic prediction based on the ITCN...
Proceedings of SPIE (December 02 2022)
An improved approach for two-stage detection model
Proceedings of SPIE (October 12 2022)
Research on text classification of telecom user query
Proceedings of SPIE (December 28 2022)
End-to-end online handwriting signature verification
Proceedings of SPIE (May 06 2019)
Event detection without trigger words on movie scripts
Proceedings of SPIE (November 10 2020)
Subscribe to Digital Library
Receive Erratum Email Alert
Back to Top