Discidius/Speech-Emotion-Classification

Full fine-tune vs facebook/wav2vec2-base-960h. Fine-tune: 95% of the weights changed, across 12 of 12 layers.

Tensors changed: Norms 52/54, Attention 96/96, MLP 48/48, Embeddings 1/1, other 2/9

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

On the Hub · Downloads on Model Pulse