Exploring End-to-End Techniques for Low-Resource Speech Recognition

TL;DR

This paper introduces a simple end-to-end grapheme-based speech recognition system for Turkish low-resource data, exploring various neural architectures and a novel CTC-loss modification, achieving state-of-the-art results.

Contribution

It presents a new end-to-end approach with a CTC-loss modification and compares multiple neural architectures for low-resource speech recognition.

Findings

Best model achieved 45.8% WER on Turkish speech data.

CTC-loss segmentation improves decoding performance.

ResNet with GRU architecture performed best among tested models.

Abstract

In this work we present simple grapheme-based system for low-resource speech recognition using Babel data for Turkish spontaneous speech (80 hours). We have investigated different neural network architectures performance, including fully-convolutional, recurrent and ResNet with GRU. Different features and normalization techniques are compared as well. We also proposed CTC-loss modification using segmentation during training, which leads to improvement while decoding with small beam size. Our best model achieved word error rate of 45.8%, which is the best reported result for end-to-end systems using in-domain data for this task, according to our knowledge.