Skip to main content
. 2017 Apr 3;50(1):323–343. doi: 10.3758/s13428-017-0873-y

Fig. 1.

Fig. 1

Three of the audio effects available in DAVID, applied on the same recording by a French female speaker, saying “Je suis en route pour la réunion” (I’m on my way to the meeting). The solid black line represents the time series of pitch values in the original recording (estimated with the SWIPE algorithm - Camacho and Harris 2008) and the red line represents the pitch of manipulated audio output. The speech waveform of the unmodified recording is shown on the x-axis of each subfigure. Pitch values on y-axis are normalized to cents with respect to mean frequency 200 Hz. a The pitch is shifted upwards by 40 cents. b Vibrato is applied with a rate of 8.5 Hz and a depth of 40 cents. c Inflection kicks in at the start of the utterance, with an initial shift of + 140 cents, and recedes after 500 ms (implemented in the happy transformation). d The three effects combined, for illustration purposes. The audio effects can be applied in any configuration