Vocal imitation of synthesised sounds varying in pitch, loudness and spectral centroid

Dataset from the vocal imitation (production) task. Includes the audio stimuli, extracted audio features (for both stimuli and imitations) and extracted parameters, along with participant metadata. Please see the paper for further details. 

Details of fields in parameter_data.csv:

Participant: index from 0-18

sex: male/female

singer: 1 if participant had been singer for > 5 years, 0 if not

feature: feature that the parameter data was extracted for

envelope: up/down for ramps, fast(5Hz)/slow(2Hz) for modulations

fail: instances where the imitation failed to meet the criteria (see paper for details)

rate: ratio of the modulation rate

extent: extent of the modulation

range: range of the ramp

slope: slope of the ramp

stimtype: type of stimulus, where single = single features, pitchamps = pitch & loudness combinations, pitchspecs = pitch & spectral centroid combinations

stimlabel: label for each stimulus. The letter indicates the feature (p=pitch, a=loudness, s=spectral centroid) and the number indicates the envelope (1 = ramp down, 2 = ramp up, 3 = 5Hz modulation, 4 = 2Hz modulation)