CAT: A CTC-CRF based ASR Toolkit Bridging the Hybrid and the End-to-end Approaches towards Data Efficiency and Low Latency
Open paper
Benchmark
Model
Rank
Results
speech-recognition-on-wsj-eval92
CTC-CRF VGG-BLSTM
#4
Word Error Rate (WER): 3.2
speech-recognition-on-aishell-1
CTC-CRF 4gram-LM
#14
Word Error Rate (WER): 6.34
Rank counts only results with a code link.