paper-with-me

홈 › Papers

LoRA-Whisper: Parameter-Efficient and Extensible Multilingual ASR

2024-06-07 · Zheshu Song, Jianheng Zhuo, Yifan Yang, Ziyang Ma, ShiXiong Zhang, Xie Chen

Recent years have witnessed significant progress in multilingual automatic speech recognition (ASR), driven by the emergence of end-to-end (E2E) models and the scaling of multilingual datasets. Despite that, two main challenges persist in multilingual ASR: language interference and the incorporation of new languages without degrading the performance of the existing ones. This paper proposes LoRA-Whisper, which incorporates LoRA matrix into Whisper for multilingual ASR, effectively mitigating language interference. Furthermore, by leveraging LoRA and the similarities between languages, we can achieve better performance on new languages while upholding consistent performance on original ones. Experiments on a real-world task across eight languages demonstrate that our proposed LoRA-Whisper yields a relative gain of 18.5% and 23.0% over the baseline system for multilingual ASR and language expansion respectively.

📄 PDF Abstract BibTeX arXiv:2406.06619

Code (0)

등록된 구현이 없습니다.

Tasks

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

Similar Papers 제목 키워드 기반

Towards Rehearsal-Free Multilingual ASR: A LoRA-based Case Study on Whisper

2024-08-20 · Tianyi Xu, Kaixun Huang, Pengcheng Guo, Yu Zhou 외

Pre-trained multilingual speech foundation models, like Whisper, have shown impressive performance across different languages. However, adapting these models to new or specific languages is computationally extensive and …

Multilingual DistilWhisper: Efficient Distillation of Multi-task Speech Models via Language-Specific Experts

2023-11-02 · Thomas Palmeira Ferraz, Marcely Zanon Boito, Caroline Brun, Vassilina Nikoulina

Whisper is a multitask and multilingual speech model covering 99 languages. It yields commendable automatic speech recognition (ASR) results in a subset of its covered languages, but the model still underperforms on a no…

Automatic Speech RecognitionAutomatic Speech Recognition (ASR)Knowledge Distillationspeech-recognition+1

Efficient Adaptation of Multilingual Models for Japanese ASR

2024-12-14 · Mark Bajo, Haruka Fukukawa, Ryuji Morita, Yuma Ogasawara

This study explores fine-tuning multilingual ASR (Automatic Speech Recognition) models, specifically OpenAI's Whisper-Tiny, to improve performance in Japanese. While multilingual models like Whisper offer versatility, th…

Automatic Speech Recognitionspeech-recognitionSpeech Recognition

Efficient Compression of Multitask Multilingual Speech Models

2024-05-02 · Thomas Palmeira Ferraz

Whisper is a multitask and multilingual speech model covering 99 languages. It yields commendable automatic speech recognition (ASR) results in a subset of its covered languages, but the model still underperforms on a no…

Automatic Speech RecognitionAutomatic Speech Recognition (ASR)Knowledge DistillationQuantization+2

Bridging the gap: A comparative exploration of Speech-LLM and end-to-end architecture for multilingual conversational ASR

2026-01-04 · Yuxiang Mei, Dongxing Xu, Jiaen Liang, Yanhua Long arxiv

The INTERSPEECH 2025 Challenge on Multilingual Conversational Speech Language Models (MLC-SLM) promotes multilingual conversational ASR with large language models (LLMs). Our previous SHNU-mASR system adopted a competiti…