search

haoheliu / voicefixer_main

General Speech Restoration
https://haoheliu.github.io/demopage-voicefixer/
MIT License
272 stars 52 forks source link

location of dataIndex/ #22

Closed xinyiyxy closed 1 month ago

xinyiyxy commented 1 month ago

@haoheliu Dear author, thank you very much for this amazing project! I'm trying to train the SSR model for dereverberation. However, I couldn't find the following files, which are mentioned in "vctk_base_ssr_unet_dereverberation.json": "data": { "source_types": ["vocals","noise"], "train_dataset":{ "vocals":{ "vctk": "dataIndex/vctk/train/speech.lst" }, "noise": { "vd_noise": "dataIndex/vd_noise/vd_noise.lst" } }, "val_dataset":{ "vocals": { "gsr": "datasets/se/TestSets/ALL_GSR/target.lst" }, "noisy": { "gsr": "datasets/se/TestSets/ALL_GSR/simulated.lst" } },

Could you help me find these files? Another question is, why are we using the "ALL_GSR" dataset here instead of "DEREV"? Thank you so much for your help and patience!

xinyiyxy commented 1 month ago

No worries I guess I've found the solution in parser. Thanks a lot.