Bandwidth Embeddings for Mixed-bandwidth Speech Recognition

TL;DR

This paper introduces a unified acoustic model for mixed-bandwidth speech recognition using bandwidth embeddings and parallel convolutional layers, improving performance on narrowband speech without harming wideband results.

Contribution

It presents a novel approach combining bandwidth embeddings and parallel convolutional layers to effectively handle mixed-bandwidth speech in a single model.

Findings

13% relative improvement on narrowband speech

Effective handling of bandwidth variability

No degradation on wideband speech

Abstract

In this paper, we tackle the problem of handling narrowband and wideband speech by building a single acoustic model (AM), also called mixed bandwidth AM. In the proposed approach, an auxiliary input feature is used to provide the bandwidth information to the model, and bandwidth embeddings are jointly learned as part of acoustic model training. Experimental evaluations show that using bandwidth embeddings helps the model to handle the variability of the narrow and wideband speech, and makes it possible to train a mixed-bandwidth AM. Furthermore, we propose to use parallel convolutional layers to handle the mismatch between the narrow and wideband speech better, where separate convolution layers are used for each type of input speech signal. Our best system achieves 13% relative improvement on narrowband speech, while not degrading on wideband speech.