A CTC Triggered Siamese Network with Spatial-Temporal Dropout for Speech Recognition

2022/06/16 by Yingying Gao, Junlan Feng, Gao, Yingying +7
Computer Science · #Audio and Speech Processing (eess.AS) #FOS: Electrical engineering #Music and Audio Processing #Speech Recognition and Synthesis #Speech and Audio Processing #electronic engineering #information engineering

paper · pdf · doi:10.48550/arxiv.2206.08031

openalex publication_date 2022/06/16 · openalex created_date 2022/06/19 · openalex updated_date 2026/07/28

Abstract

Siamese networks have shown effective results in unsupervised visual representation learning. These models are designed to learn an invariant representation of two augmentations for one input by maximizing their similarity. In this paper, we propose an effective Siamese network to improve the robustness of End-to-End automatic speech recognition (ASR). We introduce spatial-temporal dropout to support a more violent disturbance for Siamese-ASR framework. Besides, we also relax the similarity regularization to maximize the similarities of distributions on the frames that connectionist temporal classification (CTC) spikes occur rather than on all of them. The efficiency of the proposed architecture is evaluated on two benchmarks, AISHELL-1 and Librispeech, resulting in 7.13% and 6.59% relative character error rate (CER) and word error rate (WER) reductions respectively. Analysis shows that our proposed approach brings a better uniformity for the trained model and enlarges the CTC spikes obviously.

Related

• Contrastive Siamese Network for Semi-supervised Speech Recognition
• SAN: a robust end-to-end ASR model architecture
• Advances in Joint CTC-Attention based End-to-End Speech Recognition with a Deep CNN Encoder and RNN-LM
• Citrinet: Closing the Gap between Non-Autoregressive and Autoregressive End-to-End Models for Automatic Speech Recognition
• Intermediate Loss Regularization for CTC-based Speech Recognition