Continue training with trained model on custom dataset

[anilpudasaini]

Hi @lifeiteng, I am trying to fine tune a model trained on libritTTS on my custom dataset, I have prepared manifest files and generated tokens running tokenizer.py, I am stuck at a point with an error "ValueError: loaded state dict contains a parameter group that doesn't match the size of optimizer's group", how can i finetune the model if that is possible. Also can i infer without using an acoustic prompt, provided that i trained it with prefixmode=1, since i dont want to maintain speaker style?