Towards One Model to Rule All: Multilingual Strategy for Dialectal Code-Switching Arabic ASR

Shammur Absar Chowdhury, Amir Hussein, Ahmed Abdelali, Ahmed Ali
2021 Interspeech 2021   unpublished
With the advent of globalization, there is an increasing demand for multilingual automatic speech recognition (ASR), handling language and dialectal variation of spoken content. Recent studies show its efficacy over monolingual systems. In this study, we design a large multilingual end-to-end ASR using selfattention based conformer architecture. We trained the system using Arabic (Ar), English (En) and French (Fr) languages. We evaluate the system performance handling: (i) monolingual (Ar, En
more » ... d Fr); (ii) multi-dialectal (Modern Standard Arabic, along with dialectal variation such as Egyptian and Moroccan); (iii) code-switching -cross-lingual (Ar-En/Fr) and dialectal (MSA-Egyptian dialect) test cases, and compare with current state-ofthe-art systems. Furthermore, we investigate the influence of different embedding/character representations including character vs word-piece; shared vs distinct input symbol per language. Our findings demonstrate the strength of such a model by outperforming state-of-the-art monolingual dialectal Arabic and code-switching Arabic ASR. This is the first study to benchmark the performance of a multilingual ASR for dialectal and code-switching Arabic test sets. The proposed model outperforms current Arabic state-of-the-art E2E
doi:10.21437/interspeech.2021-1809 fatcat:u34wfklwmjb4fl4va5bn6gh34y