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
Thanks so much for sharing this great piece of work!
When I trained a 16k model on my own dataset, the configuration was the same as what you provided, but it was changed to my own dataset. My training results generated vertical spectra on the spectrum, as shown in the following figure. Do you know why?、
The text was updated successfully, but these errors were encountered:
Hey @zhiyu112223, this is a bit off topic. Curious how did you prepare your dataset that you used to train the 16k model, like how long in seconds where each audio files are how large was your train and val dataset. I am looking to fine tune one of the model with my own custom data and looking to see how others do it. Since the paper doesn't really dive deep on how the training data was prepare except from where the data was from.
Thanks so much for sharing this great piece of work!
When I trained a 16k model on my own dataset, the configuration was the same as what you provided, but it was changed to my own dataset. My training results generated vertical spectra on the spectrum, as shown in the following figure. Do you know why?、
The text was updated successfully, but these errors were encountered: