Towards End-to-End Speech Recognition with Deep Convolutional Neural Networks

TL;DR

This paper introduces an end-to-end speech recognition framework combining hierarchical CNNs with CTC, achieving competitive results with improved computational efficiency and the ability to model temporal correlations without RNNs.

Contribution

It proposes a novel CNN-CTC based end-to-end speech recognition model that eliminates the need for recurrent connections, enhancing efficiency and modeling capabilities.

Findings

Competitive phoneme recognition accuracy on TIMIT

Reduced computational cost compared to RNN-based models

CNNs effectively model temporal correlations with proper context

Abstract

Convolutional Neural Networks (CNNs) are effective models for reducing spectral variations and modeling spectral correlations in acoustic features for automatic speech recognition (ASR). Hybrid speech recognition systems incorporating CNNs with Hidden Markov Models/Gaussian Mixture Models (HMMs/GMMs) have achieved the state-of-the-art in various benchmarks. Meanwhile, Connectionist Temporal Classification (CTC) with Recurrent Neural Networks (RNNs), which is proposed for labeling unsegmented sequences, makes it feasible to train an end-to-end speech recognition system instead of hybrid settings. However, RNNs are computationally expensive and sometimes difficult to train. In this paper, inspired by the advantages of both CNNs and the CTC approach, we propose an end-to-end speech framework for sequence labeling, by combining hierarchical CNNs with CTC directly without recurrent connections. By evaluating the approach on the TIMIT phoneme recognition task, we show that the proposed model is not only computationally efficient, but also competitive with the existing baseline systems. Moreover, we argue that CNNs have the capability to model temporal correlations with appropriate context information.