garystafford/wav2vec2-deepfake-voice-detector

Partial fine-tune vs Gustking/wav2vec2-large-xlsr-deepfake-audio-classification. Partial change: 51% of the weights changed (attention, embeddings, mlp, norms, other), in 12 of 24 layers.

Tensors changed: other 6/20, Norms 52/114, Attention 96/192, MLP 48/96, Embeddings 4/4

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

On the Hub · Downloads on Model Pulse