Saved in:
Bibliographic Details
Main Authors: Lee, PeiYing, Guo, HauYun, Chen, Berlin
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2403.14268
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913276492775424
author Lee, PeiYing
Guo, HauYun
Chen, Berlin
author_facet Lee, PeiYing
Guo, HauYun
Chen, Berlin
contents End-to-End Neural Diarization with Encoder-Decoder based Attractor (EEND-EDA) is an end-to-end neural model for automatic speaker segmentation and labeling. It achieves the capability to handle flexible number of speakers by estimating the number of attractors. EEND-EDA, however, struggles to accurately capture local speaker dynamics. This work proposes an auxiliary loss that aims to guide the Transformer encoders at the lower layer of EEND-EDA model to enhance the effect of self-attention modules using speaker activity information. The results evaluated on public dataset Mini LibriSpeech, demonstrates the effectiveness of the work, reducing Diarization Error Rate from 30.95% to 28.17%. We will release the source code on GitHub to allow further research and reproducibility.
format Preprint
id arxiv_https___arxiv_org_abs_2403_14268
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Speech-Aware Neural Diarization with Encoder-Decoder Attractor Guided by Attention Constraints
Lee, PeiYing
Guo, HauYun
Chen, Berlin
Audio and Speech Processing
Sound
End-to-End Neural Diarization with Encoder-Decoder based Attractor (EEND-EDA) is an end-to-end neural model for automatic speaker segmentation and labeling. It achieves the capability to handle flexible number of speakers by estimating the number of attractors. EEND-EDA, however, struggles to accurately capture local speaker dynamics. This work proposes an auxiliary loss that aims to guide the Transformer encoders at the lower layer of EEND-EDA model to enhance the effect of self-attention modules using speaker activity information. The results evaluated on public dataset Mini LibriSpeech, demonstrates the effectiveness of the work, reducing Diarization Error Rate from 30.95% to 28.17%. We will release the source code on GitHub to allow further research and reproducibility.
title Speech-Aware Neural Diarization with Encoder-Decoder Attractor Guided by Attention Constraints
topic Audio and Speech Processing
Sound
url https://arxiv.org/abs/2403.14268