Geodesic interpolation of frame-wise speaker embeddings for the diarization of meeting scenarios

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Cord-Landwehr, Tobias, Boeddeker, Christoph, Zorilă, Cătălin, Doddipatla, Rama, Haeb-Umbach, Reinhold
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914634222534656
author Cord-Landwehr, Tobias
Boeddeker, Christoph
Zorilă, Cătălin
Doddipatla, Rama
Haeb-Umbach, Reinhold
author_facet Cord-Landwehr, Tobias
Boeddeker, Christoph
Zorilă, Cătălin
Doddipatla, Rama
Haeb-Umbach, Reinhold
contents We propose a modified teacher-student training for the extraction of frame-wise speaker embeddings that allows for an effective diarization of meeting scenarios containing partially overlapping speech. To this end, a geodesic distance loss is used that enforces the embeddings computed from regions with two active speakers to lie on the shortest path on a sphere between the points given by the d-vectors of each of the active speakers. Using those frame-wise speaker embeddings in clustering-based diarization outperforms segment-level clustering-based diarization systems such as VBx and Spectral Clustering. By extending our approach to a mixture-model-based diarization, the performance can be further improved, approaching the diarization error rates of diarization systems that use a dedicated overlap detection, and outperforming these systems when also employing an additional overlap detection.
format Preprint
id arxiv_https___arxiv_org_abs_2401_03963
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Geodesic interpolation of frame-wise speaker embeddings for the diarization of meeting scenarios
Cord-Landwehr, Tobias
Boeddeker, Christoph
Zorilă, Cătălin
Doddipatla, Rama
Haeb-Umbach, Reinhold
Audio and Speech Processing
We propose a modified teacher-student training for the extraction of frame-wise speaker embeddings that allows for an effective diarization of meeting scenarios containing partially overlapping speech. To this end, a geodesic distance loss is used that enforces the embeddings computed from regions with two active speakers to lie on the shortest path on a sphere between the points given by the d-vectors of each of the active speakers. Using those frame-wise speaker embeddings in clustering-based diarization outperforms segment-level clustering-based diarization systems such as VBx and Spectral Clustering. By extending our approach to a mixture-model-based diarization, the performance can be further improved, approaching the diarization error rates of diarization systems that use a dedicated overlap detection, and outperforming these systems when also employing an additional overlap detection.
title Geodesic interpolation of frame-wise speaker embeddings for the diarization of meeting scenarios
topic Audio and Speech Processing
url https://arxiv.org/abs/2401.03963