Fig. 8
From: Vocal usage learning and vocal comprehension learning in harbor seals
The 10 vocalizations of both types from seals E and J, used during training and in Experiment 1, show that the two types form two acoustically distinguishable categories. The spectrogram was obtained with the Parselmouth package in Python (v0.4.3, Praat 6.1.38; window length = 0.02s, dynamic range = 80 dB, max. frequency = 5 kHz [59,60,62])