Closed guanyuelee closed 4 years ago
Hi @guanyuelee Sorry for the late reply. I was on vacation.
I uploaded the script to store raw wav files to hdf5 (.h5) format: https://github.com/jefflai108/Contrastive-Predictive-Coding-PyTorch/blob/master/wav2raw.py
Basically, all raw wave files are stored in a single hdf5 file. The benefit of doing so is a cleaner PyTorch Dataset class code. It is my personal practice and you could, of course, do it your preferred way.
Let me know if you have further question :)
Dear Jeff,
Thank you so much for providing this great repository! Sincerely appreciate your great implementation!
However, after reading all the closed issues and trying out for initializing the training, I am still a bit confused about the training and test dataset. I try to run run.sh and the following error reported:
Traceback (most recent call last): File "main.py", line 164, in <module> main() File "main.py", line 109, in main training_set = RawDataset(args.train_raw, args.train_list, args.audio_window) File "/home/ubuntu/Contrastive-Predictive-Coding-PyTorch/src/data_reader/dataset.py", line 353, in __init__ with open(list_file) as f: FileNotFoundError: [Errno 2] No such file or directory: 'LibriSpeech/list/train.txt'
May I request what might be the possible solution of this? Thank you so much for your clarification!
Sincerely, Martin
Sorry the format is obscured. I have attached a screenshot below.
Thanks for your sharing of CPC code. I read the code and found that the provided Dataset class reads .h5 files. From open ASR website and the information provided from the paper, I can only download those files with extension .frac or .txt. Can you explicitly explain the configuration of your dataset?