No abstract

No abstract

more com r twelve call centres are planned e of the system and for consortium was oriented tors. In this view, the human CSELT, Ferrovie rafica Varese, and Saritel. ARlSE e: SNCF, IRIT, LIMSI, VECSYS KPN, KUN in Holland; PHILIPS, All the speech technology for the Italian consortium was provided by CSELT : the speech recognizer wasFlexusB and the CSELT speech synthesiser was EloquensB.

This paper presents the evaluations of the field trials of two information inquiry systems that use different speech technologies and different human-machine interfaces. The first system, RAILTEL, uses isolated word recognition and system driven dialogue. The second system, Dialogos, understands spontaneous speech and implements a mixed initiative dialogue strategy. Both the systems allow to access the Italian Railway timetable by using the telephone over the public network. Both the systems were tested in extensive field trials by inexperienced subjects. Moreover a comparative trial with a limited number of subjects was performed in order to get some insights on the impact of the use of different speech technologies on the user behaviour. After a brief description of the two systems, the paper will report the details of the experimental set up of the trials and the discussion of the results.

Two experiments are reported which were aimed at testing the effect of phonetic similarity and order of report on the right ear advantage in consonant-vowel dichotic tasks. The stimuli were consonant-vowel syllables which comprised the six stop consonants: /ba/, /da/, /ga/, /pa/, /ta/ and /ka/. In Experiment 1 the syllables were contrasted on place of articulation (condition 1) or on voicing (condition 2). In Experiment 2 the stimuli were contrasted either on one feature (place or voicing) or on both features (place and voicing). The results showed that the right ear advantage did not depend on phonetic similarity, whereas it depended on the order of report, being stronger when the perceptual channel was considered.