missing model for TSE without visual clues? #149
Description
Which model does 'Sub-Task 1: Audio-only Speaker Extraction Conditioned on a Reference Speech (8kHz)'
I could only find an old model SpEx+ that takes a reference speaker sample and extracts that speaker. But the model does not exist in the codebase anymore.