DQ-Data2vec: Decoupling Quantization for Multilingual Speech Recognition

Open in new window