Fig. 1: F0 dynamics of the word "courage" and Taylor’s (2000) tilt model.

Upper. The time course of audio signal related to the French word “courage” along with its phonetic transcription. Fricative voiceless sounds are characterized by more rapid changes with respect to voiced sound (central part of the word). For voiced sounds, the fundamental frequency (F0) can be estimated and its time contour is shown in red. Lower. Taylor’s (2000) tilt model, whereby the falling phase is present thus resulting in geometric parameters as Duration*, and Amplitude* equal to −1.