You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi,
I want to restore my speech audio samples. For that, I want to train a the voicefixer repo on my custom dataset because when I'm using the pretrained voicefixer, the voice of the speaker is getting changed slightly.
When I'm trying to train the data of custom dataset, I'm unable to do so. I'm getting the following error while training:
Runtime-error: The size of tensor a (301) must match the size of tensor b (934) at non-singleton dimension 2
This is coming during performing sanity check. The input and target tensor values are not same. Any idea how I can resolve this issue ? Also if you can mention the steps for training the model on custom data, that we be really helpful.
Thanks in advance !
Look forward to hearing from you.
Regards,
Harsh
The text was updated successfully, but these errors were encountered:
Hi,
I want to restore my speech audio samples. For that, I want to train a the voicefixer repo on my custom dataset because when I'm using the pretrained voicefixer, the voice of the speaker is getting changed slightly.
When I'm trying to train the data of custom dataset, I'm unable to do so. I'm getting the following error while training:
Runtime-error: The size of tensor a (301) must match the size of tensor b (934) at non-singleton dimension 2
This is coming during performing sanity check. The input and target tensor values are not same. Any idea how I can resolve this issue ? Also if you can mention the steps for training the model on custom data, that we be really helpful.
Thanks in advance !
Look forward to hearing from you.
Regards,
Harsh
The text was updated successfully, but these errors were encountered: