Feature Integration Strategies for Neural Speaker Diarization in Conversational Telephone Speech

Feature Integration Strategies for Neural Speaker Diarization in Conversational Telephone Speech

This paper addresses the challenge of optimizing end-to-end neural diarization systems for conversational telephone speech, focusing on diverse acoustic features beyond traditional Mel-filterbanks. We present a methodological framework for integrating and analyzing different feature types as input t...

Full description

Saved in:
Bibliographic Details
Main Authors: Juan Ignacio Alvarez-Trejos, Alicia Lozano-Diez, Daniel Ramos
Format: Article
Language:English
Published: MDPI AG 2025-04-01
Series:Applied Sciences
Subjects:
end-to-end neural diarization (EEND)
acoustic features
ECAPA-TDNN
GeMAPS
Online Access:https://www.mdpi.com/2076-3417/15/9/4842
Tags: Add Tag
No Tags, Be the first to tag this record!

Internet

https://www.mdpi.com/2076-3417/15/9/4842

Similar Items