Parameter-Efficient Adaptation of Self-Supervised Models for Arabic Speech Recognition

Read full text Summarise with VIRAT
Authors: Wafa Mohammed Alshehri, Wasfi G. Al-Khatib, Mohammad Ismail Amro
Cited by: 0
References: 12

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Arabic speech recognition systems face distinct challenges due to the language's complex morphology and dialectal variations.Self-supervised models (SSL) like XLS-R have shown promising results, but their size with over than 300 million of parameters, makes fine-tuning computationally expensive.In this work, we present the first comparative study of parameter-efficient fine-tuning (PEFT), specifically LoRA and DoRA, applied to XLS-R for Arabic ASR.We evaluate on the newly released Common Voice Arabic V24.0 dataset, establishing new benchmarks.Our full fine-tuning achieves state-of-the-art results among XLS-R-based models with 23.03% Word Error Rate (WER).In our experiments, LoRA achieved a 36.10%word error rate (WER) while training just 2% of the model's parameters.DoRA reached 45.20% WER in initial experiments.We analyze the tradeoffs between accuracy and efficiency, offering practical guidance for developing Arabic ASR systems when computational resources are limited.The models and code are publicly available.

Fields of Study & Subjects

Speech Recognition and Synthesis Speech and Audio Processing Domain Adaptation and Few-Shot Learning

Publication details

Journal / source
Workshop on NLP for Languages Using Arabic Script (AbjadNLP)
Pages
322–328
Publication date
1 January 2026
Language
EN