Single headed attention based sequence-to-sequence model for state-of-the-art results on Switchboard
Open paper
Benchmark
Model
Rank
Results
speech-recognition-on-swb-hub-500-wer
IBM (LSTM encoder-decoder)
–
Percentage error: 7.8
speech-recognition-on-switchboard-hub500
IBM (LSTM encoder-decoder)
–
Percentage error: 4.7
Rank counts only results with a code link.