search

khaotik / DaNet-Tensorflow

Tensorflow implementation of "Speaker-independent Speech Separation with Deep Attractor Network"
MIT License
89 stars 41 forks source link

the model cannot separate the mix speech! #12

Closed JayYang-Fdu closed 5 years ago

JayYang-Fdu commented 5 years ago

Hello, I am disturbing you. I used my timit dataset to find that the trained model could not separate mixed speech. May I venture to ask, do you have a trained model for reference? Thank you very much!