CAT: A CTC-CRF based ASR Toolkit Bridging the Hybrid and the End-to-end Approaches towards Data Efficiency and Low Latency

27 May 2020  ·  Keyu An, Hongyu Xiang, Zhijian Ou ·

In this paper, we present a new open source toolkit for speech recognition, named CAT (CTC-CRF based ASR Toolkit). CAT inherits the data-efficiency of the hybrid approach and the simplicity of the E2E approach, providing a full-fledged implementation of CTC-CRFs and complete training and testing scripts for a number of English and Chinese benchmarks. Experiments show CAT obtains state-of-the-art results, which are comparable to the fine-tuned hybrid models in Kaldi but with a much simpler training pipeline. Compared to existing non-modularized E2E models, CAT performs better on limited-scale datasets, demonstrating its data efficiency. Furthermore, we propose a new method called contextualized soft forgetting, which enables CAT to do streaming ASR without accuracy degradation. We hope CAT, especially the CTC-CRF based framework and software, will be of broad interest to the community, and can be further explored and improved.

PDF Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Speech Recognition AISHELL-1 CTC-CRF 4gram-LM Word Error Rate (WER) 6.34 # 10
Speech Recognition Hub5'00 FISHER-SWBD CTC-CRF Word Error Rate (WER) 12 # 1
Speech Recognition Hub5'00 SwitchBoard CTC-CRF CallHome 18.4 # 4
SwitchBoard 9.7 # 4
Hub5'00 14.1 # 1
Speech Recognition WSJ dev93 CTC-CRF VGG-BLSTM Word Error Rate (WER) 5.7 # 2
Speech Recognition WSJ eval92 CTC-CRF VGG-BLSTM Word Error Rate (WER) 3.2 # 8


No methods listed for this paper. Add relevant methods here