Why do I get particularly low indicators？

[leyangxing]

When it was in fifth epoch already overfitting。I use VCTK train_56spks dataset and test_25spks dataset. This is my trained metrics:

because in 5th epoch, it was slow change，so i only trained 10 epoch

[key2miao]

Hi, thank you very much for your questions. Please use the VCTK train_28spks to train the TSTNN. Because current version of TSTNN is low footprint. If you want to use larger dataset, please try to improve the model size like simply adding more layers, using more heads and so.

Hi, thank you very much for your questions. Please use the VCTK train_28spks to train the TSTNN. Because current version of TSTNN is low footprint. If you want to use larger dataset, please try to improve the model size like simply adding more layers, using more heads and so.