We describe here our work with automatic speech recognition (ASR) in the context of voice search functionality on the Flipkart e-Commerce platform. Starting with the deep learning architecture of Listen-Attend-Spell (LAS), we build upon and expand the model design and attention mechanisms to incorporate innovative approaches including multi-objective training, multi-pass training, and external rescoring using language models and phoneme based losses. We report a relative WER improvement of 15.7% on top of state-of-the-art LAS models using these modifications. Overall, we report an improvement of 36.9% over the phoneme-CTC system. The paper also provides an overview of different components that can be tuned in a LAS-based system.
@article{arxiv.2111.10208,
title = {Attention based end to end Speech Recognition for Voice Search in Hindi and English},
author = {Raviraj Joshi and Venkateshan Kannan},
journal= {arXiv preprint arXiv:2111.10208},
year = {2022}
}
Comments
Accepted at Forum for Information Retrieval Evaluation (FIRE) 2021