Bangla-Wave: Improving Bangla Automatic Speech Recognition Utilizing N-gram Language Models
This work addresses the lack of high-quality voice-to-text transcription for over 300 million Bangla speakers, representing an incremental improvement in a low-resource domain.
The paper tackles the problem of low-resource Bangla automatic speech recognition by fine-tuning a wav2vec2 model on the Bengali Common Voice dataset and adding an n-gram language model as a post-processor, resulting in a model that outperforms existing state-of-the-art Bengali ASR models.
Although over 300M around the world speak Bangla, scant work has been done in improving Bangla voice-to-text transcription due to Bangla being a low-resource language. However, with the introduction of the Bengali Common Voice 9.0 speech dataset, Automatic Speech Recognition (ASR) models can now be significantly improved. With 399hrs of speech recordings, Bengali Common Voice is the largest and most diversified open-source Bengali speech corpus in the world. In this paper, we outperform the SOTA pretrained Bengali ASR models by finetuning a pretrained wav2vec2 model on the common voice dataset. We also demonstrate how to significantly improve the performance of an ASR model by adding an n-gram language model as a post-processor. Finally, we do some experiments and hyperparameter tuning to generate a robust Bangla ASR model that is better than the existing ASR models.