paper-with-me

Papers

Efficient Domain Adaptation for Speech Foundation Models

2023-02-03 · Bo Li, Dongseong Hwang, Zhouyuan Huo, Junwen Bai, Guru Prakash, Tara N. Sainath, Khe Chai Sim, Yu Zhang, Wei Han, Trevor Strohman, Francoise Beaufays

Foundation models (FMs), that are trained on broad data at scale and are adaptable to a wide range of downstream tasks, have brought large interest in the research community. Benefiting from the diverse data sources such as different modalities, languages and application domains, foundation models have demonstrated strong generalization and knowledge transfer capabilities. In this paper, we present a pioneering study towards building an efficient solution for FM-based speech recognition systems. We adopt the recently developed self-supervised BEST-RQ for pretraining, and propose the joint finetuning with both source and unsupervised target domain data using JUST Hydra. The FM encoder adapter and decoder are then finetuned to the target domain with a small amount of supervised in-domain data. On a large-scale YouTube and Voice Search task, our method is shown to be both data and model parameter efficient. It achieves the same quality with only 21.6M supervised in-domain data and 130.8M finetuned parameters, compared to the 731.1M model trained from scratch on additional 300M supervised in-domain data.

📄 PDF Abstract BibTeX arXiv:2302.01496

Code (0)

등록된 구현이 없습니다.

Tasks

DecoderDomain Adaptationspeech-recognitionSpeech RecognitionTransfer Learning

Methods 이 논문이 사용한 방법론

Adapter 설명 없음
Hydra 설명 없음

Similar Papers 제목 키워드 기반

Bayesian Low-Rank Factorization for Robust Model Adaptation

2025-10-21 · Enes Yavuz Ugan, Ngoc-Quan Pham, Alexander Waibel arxiv

Large speech foundation models achieve strong performance across many domains, but they often require adaptation to handle local needs such as code-switching, where speakers mix languages within the same utterance. Direc…

SOA: Reducing Domain Mismatch in SSL Pipeline by Speech Only Adaptation for Low Resource ASR

2024-06-15 · Natarajan Balaji Shankar, Ruchao Fan, Abeer Alwan

Recently, speech foundation models have gained popularity due to their superiority in finetuning downstream ASR tasks. However, models finetuned on certain domains, such as LibriSpeech (adult read speech), behave poorly …

Domain Adaptation

Self-Taught Recognizer: Toward Unsupervised Adaptation for Speech Foundation Models

2024-05-23 · Yuchen Hu, Chen Chen, Chao-Han Huck Yang, Chengwei Qin 외

We propose an unsupervised adaptation framework, Self-TAught Recognizer (STAR), which leverages unlabeled data to enhance the robustness of automatic speech recognition (ASR) systems in diverse target domains, such as no…

Automatic Speech RecognitionAutomatic Speech Recognition (ASR)speech-recognitionSpeech Recognition

E-BATS: Efficient Backpropagation-Free Test-Time Adaptation for Speech Foundation Models

2025-06-08 · Jiaheng Dong, Hong Jia, Soumyajit Chatterjee, Abhirup Ghosh 외

Speech Foundation Models encounter significant performance degradation when deployed in real-world scenarios involving acoustic domain shifts, such as background noise and speaker accents. Test-time adaptation (TTA) has …

GPUTest-time Adaptation

SSVD: Structured SVD for Parameter-Efficient Fine-Tuning and Benchmarking under Domain Shift in ASR

2025-09-02 · Pu Wang, Shinji Watanabe, Hugo Van hamme arxiv

Parameter-efficient fine-tuning (PEFT) has emerged as a scalable solution for adapting large foundation models. While low-rank adaptation (LoRA) is widely used in speech applications, its state-of-the-art variants, e.g.,…

parameter-efficient fine-tuningSpeech RecognitionDomain Adaptation