LRS3-TED

Overview edit

Variants: LRS3-TED

Associated Benchmarks

This dataset is used in 6 benchmarks:

• Speech Recognition - Metrics: Word Error Rate (WER)
• Audio-Visual Speech Recognition - Metrics: Word Error Rate (WER)
• Lipreading - Metrics: Word Error Rate (WER)
• Visual Speech Recognition - Metrics: Word Error Rate (WER)
• Automatic Speech Recognition (ASR) - Metrics: WER, Word Error Rate (WER)
• Active Speaker Detection - Metrics: Accuracy

Recent Benchmark Submissions

Research Papers

Recent papers with results on this dataset:

• MMS-LLaMA: Efficient LLM-based Audio-Visual Speech Recognition with Minimal Multimodal Speech Tokens (2025) -
• Zero-AVSR: Zero-Shot Audio-Visual Speech Recognition with LLMs by Learning Language-Agnostic Speech Representations (2025) -
• Audio-Visual Representation Learning via Knowledge Distillation from Speech Foundation Models (2025) -
• Unified Speech Recognition: A Single Model for Auditory, Visual, and Audiovisual Inputs (2024) -
• Large Language Models are Strong Audio-Visual Speech Recognition Learners (2024) -
• SyncVSR: Data-Efficient Visual Speech Recognition with End-to-End Crossmodal Audio Token Synchronization (2024) -
• Whisper-Flamingo: Integrating Visual Features into Whisper for Audio-Visual Speech Recognition and Translation (2024) -
• Where Visual Speech Meets Language: VSP-LLM Framework for Efficient and Context-Aware Visual Speech Processing (2024) -
• GestSync: Determining who is speaking without a talking head (2023) -
• Auto-AVSR: Audio-Visual Speech Recognition with Automatic Labels (2023) -
• Conformers are All You Need for Visual Speech Recognition (2023) -
• Jointly Learning Visual and Auditory Speech Representations from Raw Data (2022) -
• Relaxed Attention for Transformer Models (2022) -
• Visual Speech Recognition for Multiple Languages in the Wild (2022) -
• Learning Audio-Visual Speech Representation by Masked Multimodal Cluster Prediction (2022) -
• Robust Self-Supervised Audio-Visual Speech Recognition (2022) -
• Sub-word Level Lip Reading With Visual Attention (2021) -
• End-to-end Audio-visual Speech Recognition with Conformers (2021) -
• Discriminative Multi-modality Speech Recognition (2020) -
• ASR is all you need: cross-modal distillation for lip reading (2019) -