LRS-3-TED

The 'LRS-3-TED' dataset is used to evaluate audio-visual multichannel speech separation and recognition systems, containing simulated multichannel overlapped speech data.

Papers using LRS-3-TED (3)

Mixed Precision DNN Qunatization for Overlapped Speech Separation and Recognition2021Audio-visual fine-tuning of audio-only ASR models2023Joint Speaker Features Learning for Audio-visual Multichannel Speech Separation and Recognition2024