{"ID":2881154,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2508.12968","arxiv_id":"2508.12968","title":"Arabic ASR on the SADA Large-Scale Arabic Speech Corpus with Transformer-Based Models","abstract":"We explore the performance of several state-of-the-art automatic speech recognition (ASR) models on a large-scale Arabic speech dataset, the SADA (Saudi Audio Dataset for Arabic), which contains 668 hours of high-quality audio from Saudi television shows. The dataset includes multiple dialects and environments, specifically a noisy subset that makes it particularly challenging for ASR. We evaluate the performance of the models on the SADA test set, and we explore the impact of fine-tuning, language models, as well as noise and denoising on their performance. We find that the best performing model is the MMS 1B model finetuned on SADA with a 4-gram language model that achieves a WER of 40.9\\% and a CER of 17.6\\% on the SADA test clean set.","short_abstract":"We explore the performance of several state-of-the-art automatic speech recognition (ASR) models on a large-scale Arabic speech dataset, the SADA (Saudi Audio Dataset for Arabic), which contains 668 hours of high-quality audio from Saudi television shows. The dataset includes multiple dialects and environments, specifi...","url_abs":"https://arxiv.org/abs/2508.12968","url_pdf":"https://arxiv.org/pdf/2508.12968v1","authors":"[\"Branislav Gerazov\",\"Marcello Politi\",\"Sébastien Bratières\"]","published":"2025-08-18T14:44:25Z","proceeding":"eess.AS","tasks":"[\"eess.AS\",\"cs.LG\"]","methods":"[\"Transformer\",\"Language Model\"]","has_code":false}