search

astorfi / 3D-convolutional-speaker-recognition

:speaker: Deep Learning & 3D Convolutional Neural Networks for Speaker Verification
Apache License 2.0
778 stars 275 forks source link

Prepare data #15

Closed TuanMang46 closed 6 years ago

TuanMang46 commented 6 years ago

Hi Astorfi, I am trying train speaker model with your model, How can I prepare my speech data (train_data, development_data, evaluation_data) for your model? Thank you very much!

chenhaishun commented 6 years ago

I have the same questions for this project. @TuanMang46
Did you solve this problem?

astorfi commented 6 years ago

Please refer to this function and read the associated paper.

Thanks, Sina

TuanMang46 commented 6 years ago

Thank for your answer, I don't know format of h5py file (train_data, development_data, evaluation_data). How can I save to h5py file. Thank you again!

astorfi commented 6 years ago

Please ignore creating HDF5 and refer to the input pipeline function provided in input_feature.py function. I modified some parts of this repository. Please make sure to have an updated fork of this repo. Thanks