bilalfaye/speecht5_tts-wolof-v0.2

Full fine-tune vs bilalfaye/speecht5_tts-wolof. Fine-tune: 100% of the weights changed, across 12 of 12 layers.

Tensors changed: Norms 111/111, other 17/17, Embeddings 3/4, Attention 192/192, MLP 72/72

1 other repos hold byte-for-byte identical weight files.

On the Hub · Downloads on Model Pulse