RahulBhalley
I think coreml conversion script takes repo string of model on HuggingFace. How do you import .ckpt file and where do u load it?
> I just download the .CPKT file and import into Mochi Duffusion using settings... MochiDiffusion's author say [here](https://github.com/godly-devotion/MochiDiffusion#models) that model should be converted to CoreML before using. I tried to...
Could you please tell: - How much RAM is needed for training? - How long does it take to train the model?
Hey @RageshAntony! Sorry I don't have answer for you (I just started exploring this repo today). I wanted to know what's the code behind https://huggingface.co/spaces/fffiloni/clone-voice-for-bark? Best, Rahul
> I renamed as "en_speaker_10.npz" and loaded as "v2/en_speaker_10", but still get "history prompt not found" error Same issue with me.
Ahh, the issue is in line 71 to 75 in generation.py. The `ALLOWED_PROMPTS` set variable restricts the name of speakers so ours is not included in it and `ValueError("history prompt...
The voice cloning is not working. :(
If you don’t care about using other speakers already present, simply use any of those names. My issue was that voice didn’t sound like it should have.
I think the correct npz file is being loaded. And the model must support every voice if it’s trained on humongous dataset like VALL-E. But I’m doubtful about the way...