Why is that when it could synthesize voice for same sampling rate data set, and next time when I reduce data size and try training it throws up dimensions mismatch error.
And for the same dat larger files get trained but duration model is not getting generated in cfg format. Throwing up errors that no such directory exist.
Why is that when it could synthesize voice for same sampling rate data set, and next time when I reduce data size and try training it throws up dimensions mismatch error.
And for the same dat larger files get trained but duration model is not getting generated in cfg format. Throwing up errors that no such directory exist.
Any help is highly appreciated. Thanks