Bengali.AI Speech Recognition: Multi-Module ASR Pipeline

2023-10-18
1 min read

From July to October 2023, I led a team of 2 in the Kaggle competition Bengali.AI Speech Recognition, and we won a silver medal, ranking 22nd of 744 teams.

Pipeline

We developed an automatic speech recognition (ASR) pipeline for a low-resource language (Bangla/Bengali). With supervised finetuning from augmented low-resource speech datasets, the inference pipeline mainly consists of:

  1. denoising;
  2. inference with the main ASR decoder model, a finetuned Wav2Vec2, whose decoder probabilities are fused with those of an n-gram draft language model;
  3. post-processing that restores the punctuation of the plain words with a BERT-based punctuation prediction model.

My part

I worked mainly on the n-gram and punctuation enhancement and on the finetuning of the Wav2Vec2 model.