Unsupervised Iterative Deep Learning of Speech Features and Acoustic Tokens with Applications to Spoken Term Detection

TL;DR

This paper introduces an unsupervised iterative deep learning framework called MATDNN that automatically discovers high-quality speech features and acoustic tokens from unlabeled data, improving spoken term detection performance.

Contribution

It proposes a novel multi-granular acoustic tokenizing deep neural network framework that iteratively optimizes speech features and tokens without labeled data.

Findings

Improved spoken term detection results on Zero Resource Speech Challenge datasets.

Effective discovery of phoneme-like tokens aligned with English phonemes.

Demonstrated iterative enhancement of speech features and tokens.

Abstract

In this paper we aim to automatically discover high quality frame-level speech features and acoustic tokens directly from unlabeled speech data. A Multi-granular Acoustic Tokenizer (MAT) was proposed for automatic discovery of multiple sets of acoustic tokens from the given corpus. Each acoustic token set is specified by a set of hyperparameters describing the model configuration. These different sets of acoustic tokens carry different characteristics for the given corpus and the language behind, thus can be mutually reinforced. The multiple sets of token labels are then used as the targets of a Multi-target Deep Neural Network (MDNN) trained on frame-level acoustic features. Bottleneck features extracted from the MDNN are then used as the feedback input to the MAT and the MDNN itself in the next iteration. The multi-granular acoustic token sets and the frame-level speech features can be iteratively optimized in the iterative deep learning framework. We call this framework the Multi-granular Acoustic Tokenizing Deep Neural Network (MATDNN). The results were evaluated using the metrics and corpora defined in the Zero Resource Speech Challenge organized at Interspeech 2015, and improved performance was obtained with a set of experiments of query-by-example spoken term detection on the same corpora. Visualization for the discovered tokens against the English phonemes was also shown.