Step 5: Usability Testing

Usability testing and tuning is an iterative process of incremental refinement of a spoken dialogue application. The goal is to improve the specifications created in step 3 and step 4. During usability testing, users interact with the spoken dialogue system as if it was already deployed. All actions are logged.

Tools for Usability Testing

Tools supplied with the Voice SDK allow to annotate the log so that problematic interactions can be identified by the application developers.

Results of the Usability Testing Process:

The result of the usability testing consists of a set of log files. The log files are similar to those generated in the data collection phase except that transcriptions from the speech recognizer are included. In addition, annotations identify those interactions that could not be handled correctly.

Supported Standards and APIs for Transcription:

EMMA

EMMA (Extensible MultiModal Annotation markup language) is used to represent the input events from the user.