Trained on a 13 min dataset on harvest with about 40% singing and 60% interviews since the lack of good samples. There is a lot of double tracked vocals too. But it turned out surprisingly good imo. Hopefully it can be of some use. I will remake this in rmvpe soon.

Model: https://huggingface.co/Albinator/Linda-McCartney-V2/resolve/main/LindaMc2_e480_s7200.zip?download=true

Samples: https://www.reddit.com/user/jetbirger5000/comments/14rrj65/linda_paul_mccartney_maybe_im_amazed_ai_cover/, https://youtu.be/E9KnPQ12Jk8?si=AJjs6LpmSBn5U-ET&t=440, https://www.weights.gg/models/clm739vww1y5icctc087fa3w5

Additional Details:

Tags: No tags available

Download Link: https://huggingface.co/Albinator/Linda-McCartney-V2/resolve/main/LindaMc2_e480_s7200.zip?download=true Hosted on Hugging Face

Convert this model in EasyAIVoice