search

NVIDIA / NeMo

A scalable generative AI framework built for researchers and developers working on Large Language Models, Multimodal, and Speech AI (Automatic Speech Recognition and Text-to-Speech)
https://docs.nvidia.com/nemo-framework/user-guide/latest/overview.html
Apache License 2.0
11.66k stars 2.44k forks source link

Error when running example training TTS on new data #2908

Closed Adibian closed 3 years ago

Adibian commented 3 years ago

Bug description

I want to train tacotron2 on new data and at first I just try to run 2_TTS_Tacotron2_Training.ipynb in colab without any changes! But when I run following code

!wget https://github.com/NVIDIA/NeMo/releases/download/v0.11.0/test_data.tar.gz && mkdir -p tests/data && tar xzf test_data.tar.gz -C tests/data

# Just like ASR, the Tacotron2 require .json files to define the training and validation data.
!cat tests/data/asr/an4_val.json

# Now that we have some sample data, we can try training Tacotron 2
# NOTE: The sample data is not enough data to properly train a Tacotron 2. This will not result in a trained Tacotron 2 and is used to illustrate how to train Tacotron 2 model
!python tacotron2.py sample_rate=16000 train_dataset=tests/data/asr/an4_train.json validation_datasets=tests/data/asr/an4_val.json trainer.max_epochs=3 trainer.accelerator=null trainer.check_val_every_n_epoch=1

I get this error:

[nltk_data] Downloading package averaged_perceptron_tagger to
[nltk_data]     /root/nltk_data...
[nltk_data]   Package averaged_perceptron_tagger is already up-to-
[nltk_data]       date!
[nltk_data] Downloading package cmudict to /root/nltk_data...
[nltk_data]   Package cmudict is already up-to-date!
[NeMo W 2021-09-27 19:13:14 experimental:28] Module <class 'nemo.collections.asr.data.audio_to_text_dali.AudioToCharDALIDataset'> is experimental, not ready for production and is not fully supported. Use at your own risk.
################################################################################
### WARNING, path does not exist: KALDI_ROOT=/mnt/matylda5/iveselyk/Tools/kaldi-trunk
###          (please add 'export KALDI_ROOT=<your_path>' in your $HOME/.profile)
###          (or run as: KALDI_ROOT=<your_path> python <your_script>.py)
################################################################################

[NeMo W 2021-09-27 19:13:14 nemo_logging:349] /usr/local/lib/python3.7/dist-packages/torchaudio/backend/utils.py:54: UserWarning: "sox" backend is being deprecated. The default backend will be changed to "sox_io" backend in 0.8.0 and "sox" backend will be removed in 0.9.0. Please migrate to "sox_io" backend. Please refer to https://github.com/pytorch/audio/issues/903 for the detail.
      '"sox" backend is being deprecated. '

Traceback (most recent call last):
  File "tacotron2.py", line 44, in <module>
    main()  # noqa pylint: disable=no-value-for-parameter
  File "/usr/local/lib/python3.7/dist-packages/nemo/core/config/hydra_runner.py", line 103, in wrapper
    strict=None,
TypeError: _run_hydra() got an unexpected keyword argument 'strict'

I changed nothing in code and just ran cells of this code that is a tutorial for training of tacotron. Thank you for any help

Adibian commented 3 years ago

I found out the problem. I was using codes in a old version.