ForeignGods/ComfyUI-Mana-Nodes

example_workflow_2.json

Tobe2d opened this issue · 5 comments

Tobe2d commented

I've updated the code and trying to test speech2text however nothing happen ;-)
the node font2img turn to red boarder and thats it.

Check the screenshot:
image

in the log it say this:

SpellChecker module is NOT accessible.
Prompt executed in 83.97 seconds

what am I missing?

ForeignGods commented

Have you tried installing this library to your python environment?

pip install pyspellchecker

I'll update the requirements.txt to include this.

Tobe2d commented

I just installed it into venv and tried again and now the log is a bit longer...

got prompt
ERROR:root:Failed to validate prompt for output 116:
ERROR:root:* font2img 113:
ERROR:root:  - Value not in list: font_file: 'Akira Expanded Demo.otf' not in (list of length 327)
ERROR:root:Output will be ignored
ERROR:root:Failed to validate prompt for output 115:
ERROR:root:Output will be ignored
ERROR:root:Failed to validate prompt for output 114:
ERROR:root:Output will be ignored
[rgthree] Using rgthree's optimized recursive execution.
MoviePy - Writing audio in D:\Ai__Project\ComfyUI\custom_nodes\ComfyUI-Mana-Nodes\audio_files\audio.wav
MoviePy - Done.
Audio file saved: D:\Ai__Project\ComfyUI\custom_nodes\ComfyUI-Mana-Nodes\audio_files\audio.wav
Full path to the audio file: D:\Ai__Project\ComfyUI\custom_nodes\ComfyUI-Mana-Nodes\audio_files\audio.wav
Some weights of the model checkpoint at ailegends/xlsr-jonatasgrosman-wav2vec2-large-xlsr-53-english were not used when initializing Wav2Vec2ForCTC: ['wav2vec2.encoder.pos_conv_embed.conv.weight_g', 'wav2vec2.encoder.pos_conv_embed.conv.weight_v']
- This IS expected if you are initializing Wav2Vec2ForCTC from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).
- This IS NOT expected if you are initializing Wav2Vec2ForCTC from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).
Some weights of Wav2Vec2ForCTC were not initialized from the model checkpoint at ailegends/xlsr-jonatasgrosman-wav2vec2-large-xlsr-53-english and are newly initialized: ['wav2vec2.encoder.pos_conv_embed.conv.parametrizations.weight.original0', 'wav2vec2.encoder.pos_conv_embed.conv.parametrizations.weight.original1']
You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.
Prompt executed in 8.47 seconds
ForeignGods commented

Now I would update to the newest version of this repo if you haven't already and then "fix node" the font2img node.

Try selecting another font.
Are the system fonts also showing?

Tobe2d commented

Yes I can see system fonts. I pulled the latest update and I changed the font to Calibri and it now it is working!!!!

ForeignGods commented

Great, I'll close this issue then.

  • 👍1