Logo Lanfrica
  • Accueil
  • Atlas
  • Analyses
  • Documentation
  • Sign in

© 2026 Lanfrica. Tous droits réservés. Tous les droits d'auteur des ressources affichées sur le site Web Lanfrica appartiennent aux détenteurs de droits d'auteur d'origine, sauf indication contraire explicite.

A Parameter-efficient Language Extension Framework for Multilingual ASR

Domaine:

natural language processing

Type de record:

papermodel
Créateur:
LiuHouYang DongCao
Hôte:avatar
Covering all languages with a multilingual speech recognition model (MASR) is very difficult. Performing language extension on top of an existing MASR is a desirable choice. In this study, the MASR continual learning problem is probabilistically decomposed into language identity prediction (LP) and cross-lingual adaptation (XLA) sub-problems. Based on this, we propose an architecture-based framework for language extension that can fundamentally solve catastrophic forgetting, debudded as PELE. PELE is designed to be parameter-efficient, incrementally incorporating an add-on module to adapt to a new language. Specifically, different parameter-efficient fine-tuning (PEFT) modules and their variants are explored as potential candidates to perform XLA. Experiments are carried out on 5 new languages with a wide range of low-resourced data sizes. The best-performing PEFT candidate can achieve satisfactory performance across all languages and demonstrates superiority in three of five languages over the continual joint learning setting. Notably, PEFT methods focusing on weight parameters or input features are revealed to be limited in performance, showing significantly inferior extension capabilities compared to inserting a lightweight module in between layers such as an Adapter. Accepted by Interspeech 2024

Visit

arxiv.org

Tasks

automatic speech recognitionspeech processing

Tags

Computation and LanguageAudio and Speech Processing

Similaires

MAPLE: Multilingual Evaluation of Parameter Efficient Finetuning of Large Language ModelsParameter-Efficient Neural Reranking for Cross-Lingual and Multilingual RetrievalBLOOM: A 176B-Parameter Open-Access Multilingual Language ModelIncreLoRA: Incremental Parameter Allocation Method for Parameter-Efficient Fine-tuningFull Fine-Tuning vs. Parameter-Efficient Adaptation for Low-Resource African ASR: A Controlled Study with Whisper-SmallLess can be More: Towards a Parameter-Efficient Fine-Tuning of Wav2Vec2 XLSR for Low-Resource Cape Verdean Creole ASR

MAPLE: Multilingual Evaluation of Parameter Efficient Finetuning of Large Language Models

Parameter Efficient Finetuning (PEFT) has emerged as a viable solution for improving the performance

Parameter-Efficient Neural Reranking for Cross-Lingual and Multilingual Retrieval

State-of-the-art neural (re)rankers are notoriously data-hungry which -- given the lack of large-sca

BLOOM: A 176B-Parameter Open-Access Multilingual Language Model

Large language models (LLMs) have been shown to be able to perform new tasks based on a few demonstr

IncreLoRA: Incremental Parameter Allocation Method for Parameter-Efficient Fine-tuning

With the increasing size of pre-trained language models (PLMs), fine-tuning all the parameters in th

Full Fine-Tuning vs. Parameter-Efficient Adaptation for Low-Resource African ASR: A Controlled Study with Whisper-Small

Less can be More: Towards a Parameter-Efficient Fine-Tuning of Wav2Vec2 XLSR for Low-Resource Cape Verdean Creole ASR