Audio Word2Vec: Unsupervised Learning of Audio Segment Representations using Sequence-to-sequence Autoencoder

TL;DR

This paper introduces Audio Word2Vec, an unsupervised method using sequence-to-sequence autoencoders to generate fixed-dimensional vector representations of variable-length audio segments, improving spoken term detection efficiency.

Contribution

It presents a novel unsupervised learning approach for audio segment embeddings using sequence-to-sequence autoencoders, outperforming traditional methods like DTW in query-by-example spoken term detection.

Findings

Significantly outperforms DTW in spoken term detection

Provides robust audio segment representations without supervision

Reduces computational requirements for audio query tasks

Abstract

The vector representations of fixed dimensionality for words (in text) offered by Word2Vec have been shown to be very useful in many application scenarios, in particular due to the semantic information they carry. This paper proposes a parallel version, the Audio Word2Vec. It offers the vector representations of fixed dimensionality for variable-length audio segments. These vector representations are shown to describe the sequential phonetic structures of the audio segments to a good degree, with very attractive real world applications such as query-by-example Spoken Term Detection (STD). In this STD application, the proposed approach significantly outperformed the conventional Dynamic Time Warping (DTW) based approaches at significantly lower computation requirements. We propose unsupervised learning of Audio Word2Vec from audio data without human annotation using Sequence-to-sequence Audoencoder (SA). SA consists of two RNNs equipped with Long Short-Term Memory (LSTM) units: the first RNN (encoder) maps the input audio sequence into a vector representation of fixed dimensionality, and the second RNN (decoder) maps the representation back to the input audio sequence. The two RNNs are jointly trained by minimizing the reconstruction error. Denoising Sequence-to-sequence Autoencoder (DSA) is furthered proposed offering more robust learning.