Bangla Voice Command Recognition in end-to-end System Using Topic Modeling based Contextual Rescoring

In this work, we perform contextual rescoring using multi-label topic modeling to improve the performance of an End-to-End Bangla voice command recognition system. We use a hybrid of Connectionist Temporal Classification (CTC) and Attention mechanism in our End-to-End architecture. We use Recurrent...

Full description

Saved in:
Bibliographic Details
Main Authors: Sadeq, Nafis, Ahmed, Shafayat, Shubha, Sudipta Saha, Islam, Md. Nahidul, Adnan, Muhammad Abdullah
Format: Conference Proceeding
Language:English
Subjects:
Context modeling
Recurrent neural networks
Resource management
Signal processing
Speech processing
Speech recognition
Task analysis
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this work, we perform contextual rescoring using multi-label topic modeling to improve the performance of an End-to-End Bangla voice command recognition system. We use a hybrid of Connectionist Temporal Classification (CTC) and Attention mechanism in our End-to-End architecture. We use Recurrent Neural Network (RNN) as language model and La-beled LDA (Latent Dirichlet allocation) for contextual rescoring. Our experiments show that our rescoring method reduces Word Error Rate (WER) from 16.7% to 12.8% in Bangla voice command recognition task when the relevant context is provided. The system does not lose any performance when irrelevant context is provided.
ISSN:2379-190X
DOI:10.1109/ICASSP40776.2020.9053970