Serhiy Stetskovych
Serhiy Stetskovych
I had the same error when using python 3.11. After using venv with python 3.10 issue has gone.
It sounds really fantastic. As I understand it can be also used with StyleTTS2? Do you have an example how it could be applied?
So as I understand the biggest problem in StyleTTS2 is vocoder? But maybe it could be replaced with codes based one?
You cannot use Indicbert tokenizer for PL-BERT because it is subword tokenizer. PL-BERT supports only word tokenizers. And it is problem for most languages differ from english. I don't know...
`tokenizer = transformers.AutoTokenizer.from_pretrained('ai4bharat/indic-bert', keep_accents=True)` tokenizer in this case will be `sentencepiece` tokenizer. Also you need to train ASR for your language because you use different language that uses different phonemes....
You didn't train `train_first.py` properly because from your logs I see that GAN train(TMA_epoch parameter in your config seems to be greater or equals to epochs_1st) has newer started. Also...