dorsal/arxiv
View SchemaDoing More with Less: Data Augmentation for Sudanese Dialect Automatic Speech Recognition
| Authors | Ayman Mansour |
|---|---|
| Categories | |
| ArXiv ID | 2601.06802vv1 |
| URL | https://arxiv.org/abs/2601.06802 |
| License | http://creativecommons.org/licenses/by/4.0/ |
Abstract
Although many Automatic Speech Recognition (ASR) systems have been developed for Modern Standard Arabic (MSA) and Dialectal Arabic (DA), few studies have focused on dialect-specific implementations, particularly for low-resource Arabic dialects such as Sudanese. This paper presents a comprehensive study of data augmentation techniques for fine-tuning OpenAI Whisper models and establishes the first benchmark for the Sudanese dialect. Two augmentation strategies are investigated: (1) self-training with pseudo-labels generated from unlabeled speech, and (2) TTS-based augmentation using synthetic speech from the Klaam TTS system. The best-performing model, Whisper-Medium fine-tuned with combined self-training and TTS augmentation (28.4 hours), achieves a Word Error Rate (WER) of 57.1% on the evaluation set and 51.6% on an out-of-domain holdout set substantially outperforming zero-shot multilingual Whisper (78.8% WER) and MSA-specialized Arabic models (73.8-123% WER). All experiments used low-cost resources (Kaggle free tier and Lightning.ai trial), demonstrating that strategic data augmentation can overcome resource limitations for low-resource dialects and provide a practical roadmap for developing ASR systems for low-resource Arabic dialects and other marginalized language varieties. The models, evaluation benchmarks, and reproducible training pipelines are publicly released to facilitate future research on low-resource Arabic ASR.
{
"annotation_id": "37b66f84-0d49-42c8-a8c3-19ba720e2f6b",
"date_created": "2026-02-17T05:53:08.813000Z",
"date_modified": "2026-02-17T05:53:08.813000Z",
"file_hash": "307ca1d2c14c464b5b097de163511d2fbb6aaa399d92b7cd16e327011852ec07",
"private": false,
"record": {
"abstract": "Although many Automatic Speech Recognition (ASR) systems have been developed for Modern Standard Arabic (MSA) and Dialectal Arabic (DA), few studies have focused on dialect-specific implementations, particularly for low-resource Arabic dialects such as Sudanese. This paper presents a comprehensive study of data augmentation techniques for fine-tuning OpenAI Whisper models and establishes the first benchmark for the Sudanese dialect. Two augmentation strategies are investigated: (1) self-training with pseudo-labels generated from unlabeled speech, and (2) TTS-based augmentation using synthetic speech from the Klaam TTS system. The best-performing model, Whisper-Medium fine-tuned with combined self-training and TTS augmentation (28.4 hours), achieves a Word Error Rate (WER) of 57.1% on the evaluation set and 51.6% on an out-of-domain holdout set substantially outperforming zero-shot multilingual Whisper (78.8% WER) and MSA-specialized Arabic models (73.8-123% WER). All experiments used low-cost resources (Kaggle free tier and Lightning.ai trial), demonstrating that strategic data augmentation can overcome resource limitations for low-resource dialects and provide a practical roadmap for developing ASR systems for low-resource Arabic dialects and other marginalized language varieties. The models, evaluation benchmarks, and reproducible training pipelines are publicly released to facilitate future research on low-resource Arabic ASR.",
"arxiv_id": "2601.06802",
"authors": [
"Ayman Mansour"
],
"categories": [
"cs.CL",
"cs.AI"
],
"license": "http://creativecommons.org/licenses/by/4.0/",
"title": "Doing More with Less: Data Augmentation for Sudanese Dialect Automatic Speech Recognition",
"url": "https://arxiv.org/abs/2601.06802",
"version": "v1"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "f6a0b3e6-a4a6-49ef-a27e-877dc2ea7780",
"id": "arXiv Dataset",
"type": "Model",
"variant": "snapshot-2026-01-17",
"version": "0.1.0"
},
"user_id": 1000002
}