Whisper in Medusa's Ear: Multi-head Efficient Decoding for Transformer-based ASR

Re-assign community