Multi-view Recurrent Neural Acoustic Word Embeddings

TL;DR

This paper introduces a multi-view deep learning approach using bidirectional LSTM models to generate fixed-dimensional acoustic word embeddings, improving word discrimination and enabling cross-view tasks.

Contribution

It presents a novel multi-view training method that jointly embeds acoustic and character sequences, enhancing speech retrieval and recognition capabilities.

Findings

Improved word discrimination over previous methods

Effective cross-view word discrimination demonstrated

Enhanced word similarity measurement

Abstract

Recent work has begun exploring neural acoustic word embeddings---fixed-dimensional vector representations of arbitrary-length speech segments corresponding to words. Such embeddings are applicable to speech retrieval and recognition tasks, where reasoning about whole words may make it possible to avoid ambiguous sub-word representations. The main idea is to map acoustic sequences to fixed-dimensional vectors such that examples of the same word are mapped to similar vectors, while different-word examples are mapped to very different vectors. In this work we take a multi-view approach to learning acoustic word embeddings, in which we jointly learn to embed acoustic sequences and their corresponding character sequences. We use deep bidirectional LSTM embedding models and multi-view contrastive losses. We study the effect of different loss variants, including fixed-margin and cost-sensitive losses. Our acoustic word embeddings improve over previous approaches for the task of word discrimination. We also present results on other tasks that are enabled by the multi-view approach, including cross-view word discrimination and word similarity.