Closed Mmiglio closed 5 years ago
Use a CNN / RNN to extract features from audio files followed by a One Class classifier trained to to recognise the keyword (marvin in our case).
marvin
As classifier we can use either a OC-SVM or Isolation Forest.
Features extracted from layer dense_1.
dense_1
Parameters of the OC-SVM:
Testsed OC classifier with all CNN models (Inception, VGG, ResNet)
Use a CNN / RNN to extract features from audio files followed by a One Class classifier trained to to recognise the keyword (
marvin
in our case).As classifier we can use either a OC-SVM or Isolation Forest.