There's No model
VITS_Luganda_Studio does not appear to have a file named pytorch_model.bin, tf_model.h5, model.ckpt or flax_model.msgpack
Excuse my delay. Didn't see the message earlier.
So, the model files are D_2800.pth (Discriminator) and G_2800.pth (Generator). This is an adversarially trained vits model which requires both discriminator and generator for training but only the generator for inference.
This colab notebook shows how to use it.
https://colab.research.google.com/drive/19f9u8h_4VDxAu0ReaFoqGAXqYlp3W7Fv?usp=sharing
I have also added a model card to make things less confusing.
Excuse my delay. Didn't see the message earlier.
So, the model files are D_2800.pth (Discriminator) and G_2800.pth (Generator). This is an adversarially trained vits model which requires both discriminator and generator for training but only the generator for inference.
This colab notebook shows how to use it.
https://colab.research.google.com/drive/19f9u8h_4VDxAu0ReaFoqGAXqYlp3W7Fv?usp=sharing
I have also added a model card to make things less confusing.
Hey. Is this dataset open sourced in raw audio clips?. I want to do some audio dialogue post production like mixing & mastering. To have an even high quality dataset.
Hi Alan,
Cool stuff you're doing,
It would be nice to have you meet the team and discuss a little more about the work you're doing.
We have fellowships open to people interested in using our models, if you're interested, see: https://sunbird.ai/careers/