| audio-visual-speech-recognition-on-lrs2 | TM-CTC | #5 | Test WER: 8.2 |
| audio-visual-speech-recognition-on-lrs2 | TM-Seq2seq | #6 | Test WER: 8.5 |
| audio-visual-speech-recognition-on-lrs3-ted | TM-seq2seq | #12 | Word Error Rate (WER): 7.2 |
| automatic-speech-recognition-on-lrs2 | TM-seq2seq | #6 | Test WER: 9.7 |
| automatic-speech-recognition-on-lrs2 | TM-CTC | #7 | Test WER: 10.1 |
| lipreading-on-lrs2 | TM-seq2seq + extLM | #11 | Word Error Rate (WER): 48.3 |
| lipreading-on-lrs2 | TM-CTC + extLM | #12 | Word Error Rate (WER): 54.7 |
| lipreading-on-lrs3-ted | TM-seq2seq | #15 | Word Error Rate (WER): 58.9 |