English

Improving Speech Recognition for Indic Languages using Language Model

Computation and Language 2022-06-16 v3 Audio and Speech Processing

Abstract

We study the effect of applying a language model (LM) on the output of Automatic Speech Recognition (ASR) systems for Indic languages. We fine-tune wav2vec 2.02.0 models for 1818 Indic languages and adjust the results with language models trained on text derived from a variety of sources. Our findings demonstrate that the average Character Error Rate (CER) decreases by over 2828 \% and the average Word Error Rate (WER) decreases by about 3636 \% after decoding with LM. We show that a large LM may not provide a substantial improvement as compared to a diverse one. We also demonstrate that high quality transcriptions can be obtained on domain-specific data without retraining the ASR model and show results on biomedical domain.

Keywords

Cite

@article{arxiv.2203.16595,
  title  = {Improving Speech Recognition for Indic Languages using Language Model},
  author = {Ankur Dhuriya and Harveen Singh Chadha and Anirudh Gupta and Priyanshi Shah and Neeraj Chhimwal and Rishabh Gaur and Vivek Raghavan},
  journal= {arXiv preprint arXiv:2203.16595},
  year   = {2022}
}

Comments

Need to upgrade the content completely

R2 v1 2026-06-24T10:32:28.629Z