I’m new to the forum, so if I’ve posted my question in the wrong section, please kindly let me know. Thank you!
My issue is with the
transformers.MusicgenProcessor for text-to-music generation. It works smoothly, but when I try to use an audio prompt, I run into a problem. The generated music has a portion at the beginning that is identical to the prompt. Why is this happening? I’m hoping for behavior similar to the demo webpage, where the prompt doesn’t appear verbatim at the start of the generated music.
Any guidance or suggestions from the experts here would be greatly appreciated. Thank you!