Saved in:
Bibliographic Details
Main Authors: Tao, Ruijie, Shi, Zhan, Jiang, Yidi, Truong, Duc-Tuan, Chng, Eng-Siong, Alioto, Massimo, Li, Haizhou
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2407.17902
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916336083402752
author Tao, Ruijie
Shi, Zhan
Jiang, Yidi
Truong, Duc-Tuan
Chng, Eng-Siong
Alioto, Massimo
Li, Haizhou
author_facet Tao, Ruijie
Shi, Zhan
Jiang, Yidi
Truong, Duc-Tuan
Chng, Eng-Siong
Alioto, Massimo
Li, Haizhou
contents The human brain has the capability to associate the unknown person's voice and face by leveraging their general relationship, referred to as ``cross-modal speaker verification''. This task poses significant challenges due to the complex relationship between the modalities. In this paper, we propose a ``Multi-stage Face-voice Association Learning with Keynote Speaker Diarization''~(MFV-KSD) framework. MFV-KSD contains a keynote speaker diarization front-end to effectively address the noisy speech inputs issue. To balance and enhance the intra-modal feature learning and inter-modal correlation understanding, MFV-KSD utilizes a novel three-stage training strategy. Our experimental results demonstrated robust performance, achieving the first rank in the 2024 Face-voice Association in Multilingual Environments (FAME) challenge with an overall Equal Error Rate (EER) of 19.9%. Details can be found in https://github.com/TaoRuijie/MFV-KSD.
format Preprint
id arxiv_https___arxiv_org_abs_2407_17902
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Multi-Stage Face-Voice Association Learning with Keynote Speaker Diarization
Tao, Ruijie
Shi, Zhan
Jiang, Yidi
Truong, Duc-Tuan
Chng, Eng-Siong
Alioto, Massimo
Li, Haizhou
Audio and Speech Processing
The human brain has the capability to associate the unknown person's voice and face by leveraging their general relationship, referred to as ``cross-modal speaker verification''. This task poses significant challenges due to the complex relationship between the modalities. In this paper, we propose a ``Multi-stage Face-voice Association Learning with Keynote Speaker Diarization''~(MFV-KSD) framework. MFV-KSD contains a keynote speaker diarization front-end to effectively address the noisy speech inputs issue. To balance and enhance the intra-modal feature learning and inter-modal correlation understanding, MFV-KSD utilizes a novel three-stage training strategy. Our experimental results demonstrated robust performance, achieving the first rank in the 2024 Face-voice Association in Multilingual Environments (FAME) challenge with an overall Equal Error Rate (EER) of 19.9%. Details can be found in https://github.com/TaoRuijie/MFV-KSD.
title Multi-Stage Face-Voice Association Learning with Keynote Speaker Diarization
topic Audio and Speech Processing
url https://arxiv.org/abs/2407.17902