efficient-speech/lite-whisper-large-v3-fast

Partial fine-tune vs openai/whisper-large-v3. Partial change: 0% of the weights changed (attention, mlp), in 21 of 32 layers.

Tensors changed: Embeddings 0/3, Attention 96/544, Norms 0/324, MLP 60/196, other 0/4

On the Hub · Downloads on Model Pulse