Skip to content submission for dcase2018 task2
Branch: master
Clone or download
Latest commit c9024e8 Sep 14, 2018
Type Name Latest commit message Commit time
Failed to load latest commit information.

dcase2018_task2_cochlearai submission for dcase2018 task2

You can download full data including models in the following links:

How to run

  1. Download the models from the above links. You can find them in '/saved_models/'.
  2. Put your audio files as 'audio_test/a.wav'.
  3. Modify the list of sample_rate_list in, as [16000], [32000], [44100], or combination of them as [16000, 44100]. (Note that those sample rates corresponds to the models, not the data. Audio files will be resampled for each sample rate models.)
  4. Run
  5. The results will be saved as submission.csv.


h5py, librosa, keras, tensorflow, kapre

You can’t perform that action at this time.