Fix automatic_model_training.ipynb for Python 3.12 / torchaudio 2.x compatibility#326
Fix automatic_model_training.ipynb for Python 3.12 / torchaudio 2.x compatibility#326Tinevagio wants to merge 1 commit intodscripka:mainfrom
Conversation
|
Hi! I spent several hours trying to get this notebook working on Google Colab This updated notebook fixes:
Tested and working end-to-end, successfully trained "hey snowy" wake word model. |
|
heyy!! how to use this fix?? |
|
m getting this error - Entraînement en cours... ❌ Erreur code 1 |
|
no worries, i built and updated the pipline |
|
hi, excuse me i was not connected
fine if you found the solution
it need to be import in google colab, and run the cells
Le dim. 5 avr. 2026 à 10:51, Anshsurana123 ***@***.***> a
écrit :
… *Anshsurana123* left a comment (dscripka/openWakeWord#326)
<#326 (comment)>
no worries, i built and updated the pipline
—
Reply to this email directly, view it on GitHub
<#326 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/B3BRQW3XSXWX2S77YHRSLT34UINBRAVCNFSM6AAAAACXMRLHL6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHM2DCOBYGU2TKNBYG4>
.
You are receiving this because you authored the thread.Message ID:
***@***.***>
|
|
Do you have a working Colab link for the fixed notebook? I've been trying to train a custom wake word all day and hitting shape mismatch errors in data.py and train.py. |
|
i open the .ipynb file in a google colab
Le lun. 6 avr. 2026 à 13:57, Anshsurana123 ***@***.***> a
écrit :
… *Anshsurana123* left a comment (dscripka/openWakeWord#326)
<#326 (comment)>
Do you have a working Colab link for the fixed notebook? I've been trying
to train a custom wake word all day and hitting shape mismatch errors in
data.py and train.py.
—
Reply to this email directly, view it on GitHub
<#326 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/B3BRQWYLYWQLL5FVBRZ6ZTT4UOLSRAVCNFSM6AAAAACXMRLHL6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHM2DCOJSGEYDMOBTGY>
.
You are receiving this because you authored the thread.Message ID:
***@***.***>
|
torchaudio.set_audio_backend deprecated
torchaudio.info removed
generate_samples.py missing
AudioSet 404 → replaced by FMA
onnxscript missing for ONNX export