search

santi-pdp / pase

Problem Agnostic Speech Encoder
MIT License
439 stars 87 forks source link

Reproduce Self-Supervised Training on LibriSpeech #108

Open Charlie0215 opened 3 years ago

Charlie0215 commented 3 years ago

I tried to reproduce the self-supervised training by pase+ on librispeech 100h training dataset. I'm just wondering, for contamination part, what is the downsample filter (downsample_irfiles in cfg file) for distortion? Do I suppose to find it in the google drive .tar file?

uuwz commented 10 months ago

Hello! I have been replicating this experiment recently, but during the process of making the dataset config file, do I know where to obtain these files. (-- train_scp data/LibriSpeed/libri_tr.scp -- test_scp data/LibriSpeed/libri_te.scp\

--Libri Dict data/LibriSpeed/Libri Dict. npy). I look forward to your reply very much. Thank you.