Saved in:
Bibliographic Details
Main Authors: Farokh, Seyed Ali, Zeinali, Hossein
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2411.10828
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914243630071808
author Farokh, Seyed Ali
Zeinali, Hossein
author_facet Farokh, Seyed Ali
Zeinali, Hossein
contents This paper presents our submission to the Iranian division of the Text-Dependent Speaker Verification Challenge (TdSV) 2024. Conventional TdSV approaches typically jointly model speaker and linguistic features, requiring unsegmented inputs during training and incurring high computational costs. Additionally, these methods often fine-tune large-scale pre-trained speaker embedding models on the target domain dataset, which may compromise the pre-trained models' original ability to capture speaker-specific characteristics. To overcome these limitations, we employ a TdSV system that utilizes two pre-trained models independently and demonstrate that, by leveraging pre-trained models with targeted domain adaptation, competitive results can be achieved while avoiding the substantial computational costs associated with joint fine-tuning on unsegmented inputs in conventional approaches. Our best system reached a MinDCF of 0.0358 on the evaluation subset and secured first place in the challenge.
format Preprint
id arxiv_https___arxiv_org_abs_2411_10828
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Memory-Efficient Training for Text-Dependent SV with Independent Pre-trained Models
Farokh, Seyed Ali
Zeinali, Hossein
Audio and Speech Processing
Computation and Language
Machine Learning
This paper presents our submission to the Iranian division of the Text-Dependent Speaker Verification Challenge (TdSV) 2024. Conventional TdSV approaches typically jointly model speaker and linguistic features, requiring unsegmented inputs during training and incurring high computational costs. Additionally, these methods often fine-tune large-scale pre-trained speaker embedding models on the target domain dataset, which may compromise the pre-trained models' original ability to capture speaker-specific characteristics. To overcome these limitations, we employ a TdSV system that utilizes two pre-trained models independently and demonstrate that, by leveraging pre-trained models with targeted domain adaptation, competitive results can be achieved while avoiding the substantial computational costs associated with joint fine-tuning on unsegmented inputs in conventional approaches. Our best system reached a MinDCF of 0.0358 on the evaluation subset and secured first place in the challenge.
title Memory-Efficient Training for Text-Dependent SV with Independent Pre-trained Models
topic Audio and Speech Processing
Computation and Language
Machine Learning
url https://arxiv.org/abs/2411.10828