I'm getting a none object when extracting glottal features from an audio file 