Saved in:
Bibliographic Details
Main Authors: Li, Xiang, Govindan, Vivek, Paturi, Rohit, Srinivasan, Sundararajan
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2406.18679
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917707022073856
author Li, Xiang
Govindan, Vivek
Paturi, Rohit
Srinivasan, Sundararajan
author_facet Li, Xiang
Govindan, Vivek
Paturi, Rohit
Srinivasan, Sundararajan
contents End-to-end neural diarization (EEND) models offer significant improvements over traditional embedding-based Speaker Diarization (SD) approaches but falls short on generalizing to long-form audio with large number of speakers. EEND-vector-clustering method mitigates this by combining local EEND with global clustering of speaker embeddings from local windows, but this requires an additional speaker embedding framework alongside the EEND module. In this paper, we propose a novel framework applying EEND both locally and globally for long-form audio without separate speaker embeddings. This approach achieves significant relative DER reduction of 13% and 10% over the conventional 1-pass EEND on Callhome American English and RT03-CTS datasets respectively and marginal improvements over EEND-vector-clustering without the need for additional speaker embeddings. Furthermore, we discuss the computational complexity of our proposed framework and explore strategies for reducing processing times.
format Preprint
id arxiv_https___arxiv_org_abs_2406_18679
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Speakers Unembedded: Embedding-free Approach to Long-form Neural Diarization
Li, Xiang
Govindan, Vivek
Paturi, Rohit
Srinivasan, Sundararajan
Audio and Speech Processing
Artificial Intelligence
Computation and Language
Machine Learning
End-to-end neural diarization (EEND) models offer significant improvements over traditional embedding-based Speaker Diarization (SD) approaches but falls short on generalizing to long-form audio with large number of speakers. EEND-vector-clustering method mitigates this by combining local EEND with global clustering of speaker embeddings from local windows, but this requires an additional speaker embedding framework alongside the EEND module. In this paper, we propose a novel framework applying EEND both locally and globally for long-form audio without separate speaker embeddings. This approach achieves significant relative DER reduction of 13% and 10% over the conventional 1-pass EEND on Callhome American English and RT03-CTS datasets respectively and marginal improvements over EEND-vector-clustering without the need for additional speaker embeddings. Furthermore, we discuss the computational complexity of our proposed framework and explore strategies for reducing processing times.
title Speakers Unembedded: Embedding-free Approach to Long-form Neural Diarization
topic Audio and Speech Processing
Artificial Intelligence
Computation and Language
Machine Learning
url https://arxiv.org/abs/2406.18679