在做data preprocessing的时候，作者提到 "During the data preprocessing phase, HuBERT [16] features and pitch contours are extracted from the audio track;". 但是似乎并没有具体说明这个pitch contours是怎么拿到的。
The authors mentioned that "During the data preprocessing phase, HuBERT [16] features and pitch contours are extracted from the audio track;" However, it appears that the authors do not explicitly present the methods they used to extract pitch contours.
在做data preprocessing的时候，作者提到 "During the data preprocessing phase, HuBERT [16] features and pitch contours are extracted from the audio track;". 但是似乎并没有具体说明这个pitch contours是怎么拿到的。
The authors mentioned that "During the data preprocessing phase, HuBERT [16] features and pitch contours are extracted from the audio track;" However, it appears that the authors do not explicitly present the methods they used to extract pitch contours.