In this paper we describe an approach to automatic evaluation of both the speech recognition and understanding capabilities of a spoken dialogue system for train time table information. We use word a c curacy for recognition and concept accuracy for understanding performance judgement. Both measures are calculated by comparing these modules' output with a… (More)
In this paper we describe the linguistic processing component of a spoken dialogue system. The task of this word graph parser is to nd the most plausible sequence of word hypotheses in the input graph. If no global solution can be found, a robust mechanism of selecting multiple partial results is applied. We argue that the semantic accuracy of the selected… (More)
This paper will describe the results of a high sophisticated speech application project. ACCeSS 1 is an EU project with Greek and German partners. Our Greek partners are Knowledge S.A. and the University of Patras. In this paper we report about the German application of the project. The project addresses a first step for automation of call centers for… (More)
Mehrdeutigkeiten sind als Lesarten in-nerhalb eines Eintrages kodiert, z.B. repräsentiert der folgende Lexikoneintrag, daß es sich bei koch entweder um einen Verbstamm oder um einen Nomenstamm im Singular handelt: Innerhalb einer Lesartliste muß das Flexionsklassenkürzel an erster Stelle stehen, danach kann eine beliebige Anzahl weiterer Merkmale folgen. 2.… (More)
This paper reports the experiences we had in evaluating the ACCeSS system using the EAGLES evaluation metrics both at the input/output (black box evaluation) and component levels (glass box evaluation). We deliver an example of a complete evaluation of a continuous speech/mixed initiative system using these standards. Furthermore, we discuss some useful… (More)
This paper presents an industrial point of view regarding dialog description languages for dialog applications on embedded devices. We present GDML, a dialog description language developed for embedding dialogs in speech control systems in cars. Furthermore we present the environment used for developing speech dialog applications with GDML. As GDML is… (More)
We describe the framework and present detailed results of an evaluation of 1.500 dialogues recorded during a three-months field-trial of the ACCeSS Dialogue System. The system was routing incoming calls to agents of a call-center and handled about 100 calls per day.