Automatic Speech Recognition on LRS3-TED (test)
7.2WERTM-seq2seq
Evaluation Results
| Method | Links | |
|---|---|---|
| TM-seq2seqModality=Audio-Visual, External Language Model=true2018.09 | 7.2 | |
| TM-CTCModality=Audio-Visual, External Language Model=true2018.09 | 7.5 | |
| TM-seq2seqModality=Audio-Visual, External Language Model=false2018.09 | 8 | |
| TM-seq2seqModality=Audio, External Language Model=true2018.09 | 8.3 | |
| TM-CTCModality=Audio, External Language Model=true2018.09 | 8.9 | |
| TM-seq2seqModality=Audio, External Language Model=false2018.09 | 9 | |
| Google S2T+Modality=Audio, External Language Model=false2018.09 | 10.4 | |
| TM-CTCModality=Audio-Visual, External Language Model=false2018.09 | 12.3 | |
| TM-CTCModality=Audio, External Language Model=false2018.09 | 13.8 | |
| TM-seq2seqModality=Video, External Language Model=true2018.09 | 58.9 | |
| TM-seq2seqModality=Video, External Language Model=false2018.09 | 59.9 | |
| TM-CTCModality=Video, External Language Model=true2018.09 | 66.3 | |
| TM-CTCModality=Video, External Language Model=false2018.09 | 74.7 |