Conv-seq2seq

Reported on 4 benchmarks across 2 tasks

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Computer Vision2 results

  • LipreadingonLRS2
    Word Error Rate (WER)· uses extra data
    51.7
    best: 14.6 (Auto-AVSR)
  • LipreadingonLRS3-TED
    Word Error Rate (WER)· uses extra data
    60.1
    best: 12.8 (LP + Conformer)

Natural Language Processing2 results