Leveraging End-to-End Speech Recognition with Neural Architecture Search

Abstract

Deep neural networks (DNNs) have been demonstrated to outperform many traditional machine learning algorithms in Automatic Speech Recognition (ASR). In this paper, we show that a large improvement in the accuracy of deep speech models can be achieved with effective Neural Architecture Optimization at a very low computational cost. Phone recognition tests with the popular LibriSpeech and TIMIT benchmarks proved this fact by displaying the ability to discover and train novel candidate models within a few hours (less than a day) many times faster than the attention-based seq2seq models. Our method achieves test error of 7% Word Error Rate (WER) on the LibriSpeech corpus and 13% Phone Error Rate (PER) on the TIMIT corpus, on par with state-of-the-art results.

arXiv commentA large part of the document needs to be reviewed to meet current standards in the Automatic Speech Recognition

References in corpus (11)

• Adam: A Method for Stochastic Optimization
• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Playing Atari with Deep Reinforcement Learning
• Neural Architecture Search with Reinforcement Learning
• On the difficulty of training Recurrent Neural Networks
• SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognition
• Deep Speech 2: End-to-End Speech Recognition in English and Mandarin
• Sequence Transduction with Recurrent Neural Networks
• Efficient Neural Architecture Search via Parameter Sharing
• Neural Architecture Optimization
• Improving Object Detection with Deep Convolutional Networks via Bayesian Optimization and Structured Prediction

Cited by in corpus (3)

• Weight-Sharing Neural Architecture Search: A Battle to Shrink the Optimization Gap
• AutoSpeech: Neural Architecture Search for Speaker Recognition
• DARTS-ASR: Differentiable Architecture Search for Multilingual Speech Recognition and Adaptation