drethage / speech-denoising-wavenet

A neural network for end-to-end speech denoising
MIT License
673 stars 165 forks source link

Is there any possibilities to change into Tensorflow model? #4

Closed tonyyyye closed 6 years ago

tonyyyye commented 6 years ago

We need to move it into Android, but there is no such interface to read from the checkpoints.

jordipons commented 6 years ago

Yes, you can change to Tensorflow since we are using Keras. You just need to switch your back-end!

As noted in the readme.md file, current code was developed under Keras 1.2 and Theano 0.9.0 - since the large dilations present in the architecture were not supported by the 1.2.0 version of Tensorflow that was available then.

Please, let us know if you find that the current version of Tensorflow (1.4) supports our large dilations!