HappyColor / SpeechFormer

Official implement of SpeechFormer written in Python (PyTorch).
75 stars 7 forks source link

Hi. Can you upload the code that generates ".csv"? #2

Closed FriedaSmith closed 1 year ago

FriedaSmith commented 2 years ago

Hi. Can you upload the code that generates the .csv?

Note that you should create a metadata file (.csv format) for each dataset to record the name and label (and state, e.g. train or dev or test) of the samples. Then modify the argument: meta_csv_file in ./config/xxx_feature_config.json according to the absolute path of the corresponding .csv file.

HappyColor commented 1 year ago

I apologize for the late reply. I have added the example .csv files in the ./metadata directory.

hvt1609 commented 1 year ago

Hi @HappyColor, I also looked at that example metadata_daicwoz_crop_resample.csv .csv. However, I still don't know how to generate it.

For example,

  | 303_s39_AUDIO.wav | 0 | 0 | train   | 303_s26_AUDIO.wav | 0 | 0 | train   | 303_s17_AUDIO.wav | 0 | 0 | train   | 303_s10_AUDIO.wav

_s39_, _s26_, _s10_ what is your mean?

Similarly,

  | 002-0-0 | Control | the scene is in the kitchen . | True   | 002-0-1 | Control | the mother is wiping dishes and the water is running on the floor . | True   | 002-0-2

Please add your code instead of. Thank you for your work,

HappyColor commented 1 year ago

I apologize for the late reply. I'm sorry that my data processing was done a long time ago and I can't find the corresponding code now. To make up for this, I have detailed how each .csvfile is generated and what they mean in the ./metadata directory.