VoxArabica: A Robust Dialect-Aware Arabic Speech Recognition System
This work addresses the problem of linguistic diversity in Arabic for researchers and users needing dialect-aware speech recognition, but it is incremental as it applies existing models like HuBERT, Whisper, and XLS-R to this domain.
The researchers tackled the challenge of building a robust automatic speech recognition (ASR) system for Arabic by developing VoxArabica, which includes dialect identification for 17 dialects plus Modern Standard Arabic and ASR models finetuned on multiple dialects, achieving a system that integrates these features into a web interface.
Arabic is a complex language with many varieties and dialects spoken by over 450 millions all around the world. Due to the linguistic diversity and variations, it is challenging to build a robust and generalized ASR system for Arabic. In this work, we address this gap by developing and demoing a system, dubbed VoxArabica, for dialect identification (DID) as well as automatic speech recognition (ASR) of Arabic. We train a wide range of models such as HuBERT (DID), Whisper, and XLS-R (ASR) in a supervised setting for Arabic DID and ASR tasks. Our DID models are trained to identify 17 different dialects in addition to MSA. We finetune our ASR models on MSA, Egyptian, Moroccan, and mixed data. Additionally, for the remaining dialects in ASR, we provide the option to choose various models such as Whisper and MMS in a zero-shot setting. We integrate these models into a single web interface with diverse features such as audio recording, file upload, model selection, and the option to raise flags for incorrect outputs. Overall, we believe VoxArabica will be useful for a wide range of audiences concerned with Arabic research. Our system is currently running at https://cdce-206-12-100-168.ngrok.io/.