Instructions to use microsoft/speecht5_tts with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use microsoft/speecht5_tts with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-to-speech", model="microsoft/speecht5_tts")# Load model directly from transformers import AutoProcessor, AutoModelForTextToSpectrogram processor = AutoProcessor.from_pretrained("microsoft/speecht5_tts") model = AutoModelForTextToSpectrogram.from_pretrained("microsoft/speecht5_tts") - Notebooks
- Google Colab
- Kaggle
Need help with cloning a new voice.
#35
by M4Marvin - opened
I am new to this, i had a doubt regarding how one would proceed if one wanted to create a clone of their own voice.
You have to fine-tune the HIFI-GAN model. Because that model generates the audio sound.
Could you please share a tutorial link with a similar model, I am inexperienced
You can see this thread I put up with a brief guide on how to do it. It's not necessary to fine-tune the HiFi-GAN model to achieve passable results, although I suspect that @imrnh 's method would give better results in the end.