Parameter-Efficient Transfer Learning of Pre-Trained Transformer Models for Speaker Verification Using Adapters

Author:

Peng Junyi1,Stafylakis Themos2,Gu Rongzhi3,Plchot Oldřich1,Mošner Ladislav1,Burget Lukáš1,Černocký Jan1

Affiliation:

1. Brno University of Technology,Faculty of Information Technology, Speech@FIT,Czechia

2. Omilia - Conversational Intelligence,Athens,Greece

3. Tencent AI Lab,Shenzhen,China

Funder

National Science Foundation

Ministry of Education

Publisher

IEEE

Reference30 articles.

1. Efficient Adapter Transfer of Self-Supervised Speech Models for Automatic Speech Recognition

2. Robust speech recognition via large-scale weak supervision;radford;Tech Rep Technical report OpenAI 2022,2022

3. Lightweight Adapter Tuning for Multilingual Speech Translation

4. Parameter-efficient transfer learning for NLP;houlsby;ICML,2019

5. Attention Back-End for Automatic Speaker Verification with Multiple Enrollment Utterances

Cited by 9 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

1. Fine-Tuning Optimization of Small Language Models: A Novel Graph-Theoretical Approach for Efficient Prompt Engineering;2024 IEEE International Symposium on Broadband Multimedia Systems and Broadcasting (BMSB);2024-06-19

2. Dual Parameter-Efficient Fine-Tuning for Speaker Representation Via Speaker Prompt Tuning and Adapters;ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP);2024-04-14

3. Leveraging in-the-wild Data for Effective Self-supervised Pretraining in Speaker Recognition;ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP);2024-04-14

4. Multiway-Adapter: Adapting Multimodal Large Language Models for Scalable Image-Text Retrieval;ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP);2024-04-14

5. Efficient Adapter Tuning of Pre-Trained Speech Models for Automatic Speaker Verification;ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP);2024-04-14

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3