EEND-M2F: Masked-attention mask transformers for speaker diarization

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Härkönen, Marc, Broughton, Samuel J., Samarakoon, Lahiru
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911763046334464
author Härkönen, Marc
Broughton, Samuel J.
Samarakoon, Lahiru
author_facet Härkönen, Marc
Broughton, Samuel J.
Samarakoon, Lahiru
contents In this paper, we make the explicit connection between image segmentation methods and end-to-end diarization methods. From these insights, we propose a novel, fully end-to-end diarization model, EEND-M2F, based on the Mask2Former architecture. Speaker representations are computed in parallel using a stack of transformer decoders, in which irrelevant frames are explicitly masked from the cross attention using predictions from previous layers. EEND-M2F is lightweight, efficient, and truly end-to-end, as it does not require any additional diarization, speaker verification, or segmentation models to run, nor does it require running any clustering algorithms. Our model achieves state-of-the-art performance on several public datasets, such as AMI, AliMeeting and RAMC. Most notably our DER of 16.07% on DIHARD-III is the first major improvement upon the challenge winning system.
format Preprint
id arxiv_https___arxiv_org_abs_2401_12600
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle EEND-M2F: Masked-attention mask transformers for speaker diarization
Härkönen, Marc
Broughton, Samuel J.
Samarakoon, Lahiru
Sound
Audio and Speech Processing
In this paper, we make the explicit connection between image segmentation methods and end-to-end diarization methods. From these insights, we propose a novel, fully end-to-end diarization model, EEND-M2F, based on the Mask2Former architecture. Speaker representations are computed in parallel using a stack of transformer decoders, in which irrelevant frames are explicitly masked from the cross attention using predictions from previous layers. EEND-M2F is lightweight, efficient, and truly end-to-end, as it does not require any additional diarization, speaker verification, or segmentation models to run, nor does it require running any clustering algorithms. Our model achieves state-of-the-art performance on several public datasets, such as AMI, AliMeeting and RAMC. Most notably our DER of 16.07% on DIHARD-III is the first major improvement upon the challenge winning system.
title EEND-M2F: Masked-attention mask transformers for speaker diarization
topic Sound
Audio and Speech Processing
url https://arxiv.org/abs/2401.12600