quick question
are you able to make a tutorial on how to use the models? like using the audios and stuff like that..
because when i open the colab im pretty confused on what to do.
Go ahead check the https://huggingface.co/spaces/ArkanDash/rvc-models-new
I dont have plan to make a tutorial maybe soon...
To use the model just run the cell
- Install Dependencies
- Run as gradio
But if you want to install custom model just create folder with the model you've added inside the weight folder with model info (if you turn on the model info checkbox), if you dont want some complicated stuff with the model info. You can just turn it off. Basically it will read the file inside every single folder that has pth and index file. No total_fea.npy required
Why do the new models that have been uploaded have faster processing?
What do you mean by that?
I'm not sure if it's just my device and/or internet, but the new models like Kaeya process songs much faster. It's absolutely amazing!
I think you might referring to v2 model, but I dont quite understand yet. On the original rvc webui doesn't have options for v1 or v2 models. But after look some issues in the main repo. There is someone mentioning v2 and v1 model.
I might retrain older model soon...
What next after installing the dependencies and running the gradio?, I'm confused after opening the public URL and I don't know what to do, pls
You can use the youtube link to separate audio and music from music video (any kind of video), the audio will be put on the audio path in the middle one and then click generate
or
You can click the "upload mode" to upload your own audio file and then click generate
Were having updates right now. Please wait
How do I create my own model, I already have .pth, .jpg files but I don't know how to get total_fea.npy file, or added_IVF415_Flat_nprobe_6.index file