Saved in:
Bibliographic Details
Main Authors: Kan, Xuan, Xiao, Yonghui, Yang, Tien-Ju, Chen, Nanxin, Mathews, Rajiv
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2408.11873
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911999051431936
author Kan, Xuan
Xiao, Yonghui
Yang, Tien-Ju
Chen, Nanxin
Mathews, Rajiv
author_facet Kan, Xuan
Xiao, Yonghui
Yang, Tien-Ju
Chen, Nanxin
Mathews, Rajiv
contents This work explores the challenge of enhancing Automatic Speech Recognition (ASR) model performance across various user-specific domains while preserving user data privacy. We employ federated learning and parameter-efficient domain adaptation methods to solve the (1) massive data requirement of ASR models from user-specific scenarios and (2) the substantial communication cost between servers and clients during federated learning. We demonstrate that when equipped with proper adapters, ASR models under federated tuning can achieve similar performance compared with centralized tuning ones, thus providing a potential direction for future privacy-preserved ASR services. Besides, we investigate the efficiency of different adapters and adapter incorporation strategies under the federated learning setting.
format Preprint
id arxiv_https___arxiv_org_abs_2408_11873
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Parameter-Efficient Transfer Learning under Federated Learning for Automatic Speech Recognition
Kan, Xuan
Xiao, Yonghui
Yang, Tien-Ju
Chen, Nanxin
Mathews, Rajiv
Audio and Speech Processing
Cryptography and Security
Machine Learning
This work explores the challenge of enhancing Automatic Speech Recognition (ASR) model performance across various user-specific domains while preserving user data privacy. We employ federated learning and parameter-efficient domain adaptation methods to solve the (1) massive data requirement of ASR models from user-specific scenarios and (2) the substantial communication cost between servers and clients during federated learning. We demonstrate that when equipped with proper adapters, ASR models under federated tuning can achieve similar performance compared with centralized tuning ones, thus providing a potential direction for future privacy-preserved ASR services. Besides, we investigate the efficiency of different adapters and adapter incorporation strategies under the federated learning setting.
title Parameter-Efficient Transfer Learning under Federated Learning for Automatic Speech Recognition
topic Audio and Speech Processing
Cryptography and Security
Machine Learning
url https://arxiv.org/abs/2408.11873