HyperAIHyperAI
HomeNewsPapersTutorialsDatasetsWikiSOTALLM ModelsGPU LeaderboardEvents
Search
About
HyperAIHyperAI

Audio Visual Speech Recognition On Lrs3 Ted

Metrics

Word Error Rate (WER)

Results

Performance results of various models on this benchmark

Model Name
Word Error Rate (WER)
Paper TitleRepository
EG-seq2seq6.8Discriminative Multi-modality Speech Recognition
DistillAV1.3Audio-Visual Representation Learning via Knowledge Distillation from Speech Foundation Models
TM-seq2seq7.2Deep Audio-Visual Speech Recognition
RNN-T4.5Recurrent Neural Network Transducer for Audio-Visual Speech Recognition
Hyb-Conformer2.3End-to-end Audio-visual Speech Recognition with Conformers
AV-HuBERT Large1.4Robust Self-Supervised Audio-Visual Speech Recognition
Llama-AVSR0.77Large Language Models are Strong Audio-Visual Speech Recognition Learners
CTC/Attention0.9Auto-AVSR: Audio-Visual Speech Recognition with Automatic Labels
Whisper-Flamingo0.76Whisper-Flamingo: Integrating Visual Features into Whisper for Audio-Visual Speech Recognition and Translation
RAVEn Large1.4Jointly Learning Visual and Auditory Speech Representations from Raw Data
Zero-AVSR1.5Zero-AVSR: Zero-Shot Audio-Visual Speech Recognition with LLMs by Learning Language-Agnostic Speech Representations
MMS-LLaMA0.74MMS-LLaMA: Efficient LLM-based Audio-Visual Speech Recognition with Minimal Multimodal Speech Tokens
0 of 12 row(s) selected.