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
Currently it is possible to mix num_windows and has_dynamic_features between the acoustic model and feature generation configuration.
Depending on the model that is being trained the result will be different. Some will train but perform poorly and others will fail to train with an arbitrary error.
Adding an warning for when the settings do not match may reduce risk of user-error.
The text was updated successfully, but these errors were encountered:
SouperDuper
changed the title
[suggeston] Add an exception for acoustic model config not matching feature generation
[suggeston] Add an warning for acoustic model config not matching feature generation
Dec 2, 2022
Currently it is possible to mix
num_windows
andhas_dynamic_features
between the acoustic model and feature generation configuration.Depending on the model that is being trained the result will be different. Some will train but perform poorly and others will fail to train with an arbitrary error.
Adding an warning for when the settings do not match may reduce risk of user-error.
The text was updated successfully, but these errors were encountered: