| license: apache-2.0 | |
| The AV_MossFormer2_TSE_16K model weights for 16 kHz audio-visual target speaker extraction in [ClearerVoice-Studio](https://github.com/modelscope/ClearerVoice-Studio/tree/main) repo. | |
| This model is trained on large scale open-sourced datasets. | |
| It extracts each speaker's voice from a multi-speaker video using facial recognition. |