https://uberduck.ai/ logo
Powered by
# machine-learning
  • l
    oh go into images
  • c
    Right now it sounds like some of the sample lines followed by garbage data
  • c
    downloading the audio downloads an htm with "no file"
  • c
    That quick beige fox jumped in the air loudly over the thin dog fence.
  • c
    the graph looks like this but the audio doesn't sound too good
  • c
    should i continue
  • c
    and why is the audio wonky anyway
  • c
    Already tried this
  • c
    didn't work and got the same error
  • c
    I may end up submitting it completely untested
  • c
    even using the correct notebook
  • c
    ANOTHER TRACEBACK
  • c
    ANY help please
  • l
    back
  • l
    im back from taking a shower so lemme help you
  • l
    so restart the training thing
  • c
    I just finalized this model
  • l
    UGHHH
  • c
    I want to synthesize it
  • l
    you can't synthesize it you can try by asking pikachu
  • c
    why not
  • c
    I have the voice file
  • c
    It's already trained for 100
  • c
    Pikachu is offline as well
  • l
    for the batch size you can either do 1 or 4 or 3 idek
  • c
    I kept the default
  • l
    you cant bc idk i tried synthesizing it but you cant synthesize it
  • c
    so I guess I trained this model for nothing
  • l
    if you want ur model to be high quality copy these settings
  • c
    oh
1...922923924...1068Latest