This paper presents the qTA toolkit, a general-purpose research toolkit for studying speech prosody. The toolkit consists of analysis and visualization tools. The analysis tool processes F0 and timing data together with annotation of communicative functions to estimate function-specific underlying pitch targets and their function-specific adjustments. The visualization tool generates illustrations of the synthesized F0 contour and their pitch target input. As an initial test, the qTA toolkit is applied to a Mandarin corpus, and the results suggest that it can be effectively used for investigating prosody in terms of communicative functions.
Index Terms: quantitative target approximation, qTA model, pitch target, communicative function, research toolkit
Cite as: Prom-on, S., Xu, Y. (2010) The qTA toolkit for prosody: learning underlying parameters of communicative functions through modeling. Proc. Speech Prosody 2010, paper 034
@inproceedings{promon10_speechprosody,
author={Santitham Prom-on and Yi Xu},
title={{The qTA toolkit for prosody: learning underlying parameters of communicative functions through modeling}},
year=2010,
booktitle={Proc. Speech Prosody 2010},
pages={paper 034}
}