search

kan-bayashi / ParallelWaveGAN

Unofficial Parallel WaveGAN (+ MelGAN & Multi-band MelGAN & HiFi-GAN & StyleMelGAN) with Pytorch
https://kan-bayashi.github.io/ParallelWaveGAN/
MIT License
1.57k stars 343 forks source link

Normalization fails on own dataset #341

Closed prajwaljpj closed 2 years ago

prajwaljpj commented 2 years ago

Using own dataset for training multiband melgan. when running 

./run.sh --conf conf/multi_band_melgan.v2.yaml --stage 1

I get this error:

(venv) ubuntu@tts-server-base-gpu-2:~/projects/ParallelWaveGAN/egs/ljspeech/voc1$ ./run.sh --conf conf/multi_band_melgan.v2.yaml --stage 1
Stage 1: Feature extraction
Feature extraction start. See the progress via dump/train_nodev/raw/preprocessing.*.log.
Feature extraction start. See the progress via dump/dev/raw/preprocessing.*.log.
Feature extraction start. See the progress via dump/eval/raw/preprocessing.*.log.
Successfully make subsets.
Successfully make subsets.
Successfully make subsets.
Successfully finished feature extraction of eval set.
Successfully finished feature extraction of dev set.
Successfully finished feature extraction of train_nodev set.
Successfully finished feature extraction.
Statistics computation start. See the progress via dump/train_nodev/compute_statistics.log.
Successfully finished calculation of statistics.
Nomalization start. See the progress via dump/train_nodev/norm/normalize.*.log.
Nomalization start. See the progress via dump/dev/norm/normalize.*.log.
Nomalization start. See the progress via dump/eval/norm/normalize.*.log.
run.pl: 7 / 104 failed, log is in dump/dev/norm/normalize.*.log
run.pl: 23 / 104 failed, log is in dump/eval/norm/normalize.*.log
run.pl: 1 / 104 failed, log is in dump/train_nodev/norm/normalize.*.log
./run.sh: 3 background jobs are failed.

Looking into the log files there are these errors for some files

Traceback (most recent call last):
  File "/home/ubuntu/projects/espnet/tools/venv/bin/parallel-wavegan-normalize", line 11, in <module>
    load_entry_point('parallel-wavegan', 'console_scripts', 'parallel-wavegan-normalize')()
  File "/home/ubuntu/projects/ParallelWaveGAN/parallel_wavegan/bin/normalize.py", line 134, in main
    dataset = AudioMelDataset(
  File "/home/ubuntu/projects/ParallelWaveGAN/parallel_wavegan/datasets/audio_mel_dataset.py", line 85, in __init__
    assert len(audio_files) != 0, f"Not found any audio files in ${root_dir}."
AssertionError: Not found any audio files in $dump/dev/raw/dump.99.
# Accounting: time=7 threads=1
# Ended (code 1) at Tue Mar 15 15:15:13 UTC 2022, elapsed time 7 seconds

Have I missed any steps?

prajwaljpj commented 2 years ago

I had changed the n_jobs=104 and n_jobs=64 which did not work. Using default n_jobs=16 worked! Closing