Faster MusicGen Generation with Streaming

#23
by sanchit-gandhi HF staff - opened

There's no need to wait for MusicGen to generate the full audio before you can start listening to the outputs ⏰ With streaming, you can play the audio as soon as the first chunk is ready 🎵 In practice, this reduces the latency to just 5s ⚡️

Check-out the demo: https://huggingface.co/spaces/sanchit-gandhi/musicgen-streaming

It's not working to download the audio with your link.

Can someone who does prompt engineering work help me? With what formulation can we get the best results from the model, what kind of prompt should we give that will give us the most compatible and best quality sound with our prompt?

Sign up or log in to comment