LRS-3-TED

The LRS3-TED dataset/benchmark contains audio-visual recordings of speech and is used to evaluate the performance of speech separation and recognition systems.

Papers using LRS-3-TED (3)

Joint Speaker Features Learning For Audio-visual Multichannel Speech Separation And Recognition2024Audio-visual fine-tuning of audio-only ASR models2023Joint Speaker Features Learning for Audio-visual Multichannel Speech Separation and Recognition2024