TasksSotADatasetsPapersMethodsSubmitAbout
Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable BenchmarksAll SotADatasetsPapersMethods

Community

Submit ResultsAbout

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

Models/CTC/Attention

CTC/Attention

Reported on 11 benchmarks across 6 tasks · 3 papers · 10 SOTA

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

Speech6 results

  • Audio-Visual Speech RecognitiononLRS3-TED
    Word Error Rate (WER)· uses extra data· 2023-03-25
    0.9
    best: 0.74 (MMS-LLaMA)
    SOTA
    Auto-AVSR: Audio-Visual Speech Recognition with Automatic LabelsarXiv:2303.14307
  • Audio-Visual Speech RecognitiononLRS2
    Test WER· uses extra data· 2023-03-25
    1.5
    best: 1.4 (Whisper-Flamingo)
    SOTA
    Auto-AVSR: Audio-Visual Speech Recognition with Automatic LabelsarXiv:2303.14307
  • Visual Speech RecognitiononLRS3-TED
    Word Error Rate (WER)· uses extra data· 2023-03-25
    19.1
    SOTA
    Auto-AVSR: Audio-Visual Speech Recognition with Automatic LabelsarXiv:2303.14307
  • Automatic Speech Recognition (ASR)onLRS2
    Test WER· uses extra data· 2023-03-25
    1.5
    best: 1.3 (Whisper)
    SOTA
    Auto-AVSR: Audio-Visual Speech Recognition with Automatic LabelsarXiv:2303.14307
  • Automatic Speech Recognition (ASR)onLRS3-TED
    Word Error Rate (WER)· 2023-03-25
    1
    SOTA
    Auto-AVSR: Audio-Visual Speech Recognition with Automatic LabelsarXiv:2303.14307
  • Audio-Visual Speech RecognitiononLRS2
    Test WER· 2018-09-28
    7
    best: 1.4 (Whisper-Flamingo)
    SOTA
    Audio-Visual Speech Recognition With A Hybrid CTC/Attention ArchitecturearXiv:1810.00108

Audio3 results

  • Speech RecognitiononLRS2
    Test WER· uses extra data· 2023-03-25
    1.5
    best: 1.3 (Whisper)
    SOTA
    Auto-AVSR: Audio-Visual Speech Recognition with Automatic LabelsarXiv:2303.14307
  • Speech RecognitiononLRS3-TED
    Word Error Rate (WER)· 2023-03-25
    1
    best: 0.68 (Whisper)
    SOTA
    Auto-AVSR: Audio-Visual Speech Recognition with Automatic LabelsarXiv:2303.14307
  • Speech RecognitiononLRS3-TED
    Word Error Rate (WER)· uses extra data· 2023-03-25
    19.1
    best: 0.68 (Whisper)
    Auto-AVSR: Audio-Visual Speech Recognition with Automatic LabelsarXiv:2303.14307

Computer Vision2 results

  • LipreadingonGRID corpus (mixed-speech)
    Word Error Rate (WER)· uses extra data· 2022-02-26
    1.2
    SOTA
    Visual Speech Recognition for Multiple Languages in the WildarXiv:2202.13084
  • LipreadingonLRS2
    Word Error Rate (WER)· 2022-02-26
    32.9
    best: 14.6 (Auto-AVSR)
    Visual Speech Recognition for Multiple Languages in the WildarXiv:2202.13084

Natural Language Processing2 results

  • Natural Language TransductiononGRID corpus (mixed-speech)
    Word Error Rate (WER)· uses extra data· 2022-02-26
    1.2
    SOTA
    Visual Speech Recognition for Multiple Languages in the WildarXiv:2202.13084
  • Natural Language TransductiononLRS2
    Word Error Rate (WER)· 2022-02-26
    32.9
    best: 14.6 (Auto-AVSR)
    Visual Speech Recognition for Multiple Languages in the WildarXiv:2202.13084