TypeError: guvectorize() missing 1 required positional argument 'signature'

NVIDIA / flowtron

Flowtron is an auto-regressive flow-based generative network for text to speech synthesis with control over speech variation and style transfer

https://nv-adlr.github.io/Flowtron

Apache License 2.0

887 stars 177 forks source link

Open nafonelli opened 2 years ago

nafonelli commented 2 years ago

When training from scratch using: python3 train.py -c config.json -p train_config.output_directory=outdir data_config.use_attn_prior=1

I get the following error: TypeError: guvectorize() missing 1 required positional argument 'signature'

I'm using the LJSpeech dataset, the traceback starts in 'train.py' on line 27 which appears to be when it imports data and datacollate

Apologies if this is trivial, I'm new to this!

Many thanks for any suggestions

egorsmkv commented 1 year ago

Solved like this:

pip install resampy==0.3.1