Finetune for single speaker
Hi @ylacombe
I'm trying to fine-tune the Indic Parler TTS model for a single speaker using this notebook: https://colab.research.google.com/github/ylacombe/scripts_and_notebooks/blob/main/Finetuning_Parler_TTS_on_a_single_speaker_dataset.ipynb
I replaced:
The model path with "ai4bharat/indic-parler-tts-pretrained".
The prompt and description tokenizer with "google/flan-t5-large".
The feature extractor with "ylacombe/dac_44khz".
However, I’m encountering this error:TypeError: dacmodel.encode() got an unexpected keyword argument 'bandwidth'.
Your assistance would mean a lot to me. Thanks in advance!
Hi @ylacombe and @AshwinSankar ,
I'm also facing the same issue. It would be really helpful if either of you could update the notebook to support fine-tuning the Indic Parler TTS model. Thank you so much!