Issue in scoring on different .wav sound files

MiteshPuthran / Speech-Emotion-Analyzer

The neural network model is capable of detecting five different male/female emotions from audio speeches. (Deep Learning, NLP, Python)

MIT License

1.3k stars 438 forks source link

Issue in scoring on different .wav sound files #24

Closed yashkumar94 closed 5 years ago

yashkumar94 commented 5 years ago

Whenever i am scoring the model on different set of .wav sound, during extracting the features with same parameters I am getting different number of features. How to overcome this problem?

MiteshPuthran commented 5 years ago

The reason is you are just relying on the parameters form the code file. The audio that you provide should be clipped to 4 sec. You can find more parameters in the audio recorder notebook file.

Following that you can find same number of features after it's extracted.