Listening test GUI to label speech files (generally wav, monospeaker) according to perceived speaker characteristics, such as likable, attractive, competent, childish, etc. Labels and pop ups for the listener are writen in German.
A semantic-differential questionnaire of 34 items is presented to the test participants. Each item consists on a pair of antonyms related to a speaker characteritic (e.g. likable / non-likable, insecure / secure,..). For each stimulus file, the listener is asked to select a value on continuous sliders corresponding to each questionnaire item. The stimuli are taken from a given folder and are randomly played.
Speech test stimulus files are to be placed in the corresponding male or female folder of the project structure.
This questionnaire was used to label the Nautilus Speaker Characterization (NSC) Corpus.
More information about the questionnaire items in:
Fernández Gallardo, L. and Weiss, B., "The Nautilus Speaker Characterization Corpus: Speech Recordings and Labels of Speaker Characteristics and Voice Descriptions," submitted to International Conference on Language Resources and Evaluation (LREC), 2018.
I am very open for comments, amends, contributions,..