Fresh meat here. Just installed. Never used but won't pay monthly for what I can get for free

Custom Nodes
1

Never used comfy and only installed it yesterday. I thought I’d start with something simple (the voice). I have an MP3 voice I want to clone to use in my videos as my voice. I installed comfy and the Higgs V3 TTS suite. Grabbed a workflow (the things you hook the wires to) from the internet and dropped it in the program. Replace the default voice with my MP3 and clicked run. Much to my surprise … it failed. Error first node.

Error

# ComfyUI Error Report

Error Details

  • **Node ID:** 1
  • **Node Type:** HiggsV3LoadModel
  • **Exception Type:** Exception
  • **Exception Message:** Exception: expected value at line 1 column 1

File “C:\Users\myname\AppData\Local\Comfy-Desktop\ComfyUI-Installs\BRZRKR\ComfyUI\execution.py”, line 545, in execute
output_data, output_ui, has_subgraph, has_pending_tasks = await get_output_data(prompt_id, unique_id, obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, v3_data=v3_data)

I have no idea what all that means. I just want a simple voice .. help.

2

OK, got a voice to work with omnivoice but that doesn’t tell me what’s wrong with the Higgs suite. I’m sure I’m like most people and would like to try different suites. But it’s something I can work with now.

On to avatar.

3

What I’ve learnt on this AI video generation quest.

1. AI isn’t so smart after all.

2. Just because someone gives you a workflow/layout doesn’t mean it’s going to work.

  1. They trick you into downloading GB’s of data that you don’t need.
  2. Every tutorial version is different. Different python, different comfi … etc.
  3. I have to hunt down all the files I need to run that workflow and place them in folders manually.
  4. It’s AI, why when I open a workflow doesn’t it tell me what I’m missing to make this workflow work and install it itself.