https://uberduck.ai/ logo
Join Discord
Powered by
# tacotron-2-support
  • g

    Gosmokeless28

    01/30/2023, 1:43 AM
    Huh?
  • t

    TwoTrickOmnic

    01/30/2023, 1:45 AM
    /content/project --------------------------------------------------------------------------- ValueError Traceback (most recent call last) in 171 with open(target_filelist) as f: 172 lines = f.read().splitlines() --> 173 speaker_count = max([int(i.split("|")[2]) for i in lines]) + 1 174 wav_count = len(lines) 175 ValueError: max() arg is an empty sequence
  • t

    TwoTrickOmnic

    01/30/2023, 1:45 AM
    i think it has something to do with the base model
  • t

    TwoTrickOmnic

    01/30/2023, 1:48 AM
    but hey what do i know i am a machine learning peabrain
  • t

    TwoTrickOmnic

    01/30/2023, 2:05 AM
    STATUS UPDATE: NEVERMIND ITS WORKING!!
  • g

    Gosmokeless28

    01/30/2023, 2:09 AM
    Yay
  • t

    TwoTrickOmnic

    01/30/2023, 2:10 AM
    wait
  • t

    TwoTrickOmnic

    01/30/2023, 2:10 AM
    is training SUPPOSED to run for only 11 seconds?
  • g

    Gosmokeless28

    01/30/2023, 2:41 AM
    No
  • a

    AhmadGT

    01/30/2023, 3:27 PM
    what's the best batch size to train a model?
  • c

    Cris140

    01/30/2023, 4:51 PM
    Depends on your dataset
  • c

    Cris140

    01/30/2023, 4:51 PM
    Less data means lower batch size
  • a

    AhmadGT

    01/30/2023, 5:00 PM
    i have 37 wavs and i think 6 or 5 minutes of voice data
  • a

    AhmadGT

    01/30/2023, 5:01 PM
    is 6 batch size good?
  • c

    Cris140

    01/30/2023, 5:03 PM
    10~14 should be fine for it
  • a

    AhmadGT

    01/30/2023, 5:04 PM
    okay
  • a

    AhmadGT

    01/30/2023, 5:04 PM
    but if I used 6 will be a problem?
  • c

    Cris140

    01/30/2023, 5:05 PM
    It will take longer to train
  • a

    AhmadGT

    01/30/2023, 11:48 PM
    Synthesis notebook no work
  • u

    {K EY1} (Kei)

    01/31/2023, 12:13 AM
    Send the error
  • t

    TwoTrickOmnic

    01/31/2023, 11:58 AM
    okayyyyy
  • t

    TwoTrickOmnic

    01/31/2023, 11:58 AM
    another problem
  • t

    TwoTrickOmnic

    01/31/2023, 11:59 AM
    so the training only runs for 11 seconds and outputs an error without the little play button turning red the error is: FileNotFoundError: [Errno 2] No such file or directory: 'speakers/0000_Ramattra/Ramattra/00000005D9C0.0B2-Yes.wav'
  • t

    TwoTrickOmnic

    01/31/2023, 11:59 AM
    but i CLEARLY have that directory set up
  • a

    AhmadGT

    01/31/2023, 1:41 PM
    fatal: destination path 'hifi-gan' already exists and is not an empty directory.
  • e

    Epic toad

    01/31/2023, 1:41 PM
    Training isn’t meant to run 11 seconds
  • a

    AhmadGT

    01/31/2023, 1:46 PM
    i think he mean the training cell runs for a little bit and then stops
  • d

    Dyno

    01/31/2023, 1:46 PM
    If you have trouble using any notebooks, make sure you use the correct notebooks, which are linked in the #notebooks channel, as those notebooks are the most up-to-date ones.
  • a

    AhmadGT

    01/31/2023, 1:46 PM
    I D I D
  • a

    AhmadGT

    01/31/2023, 1:49 PM
    it does that when not gpu
1...119120121...158Latest