You are here
DICIT Acoustic WOZ Data
FBK-irst has conducted Wizard of Oz experiments with the aim of collecting useful data for testing signal processing algorithms in the scenario foreseen by the DICIT project.
Usually, in a Wizard of Oz experiment, a subject is requested to complete specific tasks using an artificial system. The user is told that the system is fully functional and should try to use it in an intuitive way, while the system is operated by a person not visible to the subject. The operating person - called wizard - can react to user inputs in a more comprehensive way than any system could, because he/she is not confined by computer logic.
In an effort to simulate as closely as possible the behaviour of a real system based on voice interaction, recognition errors are randomly simulated by the wizard.
The goal of our WOZ experiments is to create realistic usage scenarios for acoustic pre-processing purposes. As a consequence our interest in a high level analysis of the interaction is limited. This data can be useful for testing in a real and adverse scenario different types of signal processing algorithms: Acoustic Echo Cancellation (AEC), Speech Activity Detection (SAD), Sound Source Localization (SLoc), Speaker Identification.