Abstract The commercial viability of automating large scale directory assistance is shown by presenting new results on the recognition of large numbers of different names. Satisfactory recognition performance is achieved by employing a stochastic combination of N-best lists retrieved from multiple user utterances with the telephone database as an additional knowledge source. The strategy is used in a prototype of a fully automated directory information system which is designed to cover a whole country: After the city has been selected, the user is asked for first and last name of the desired person and, if necessary, also for the street or a spelling of the last name. Confidence measures are used for an optimal dialogue flow.

This paper describes a design and feasibility study for a large-scale automatic directory information system with a scalable architecture. The current demonstrator, called PADIS-XL 1, operates in realtime and handles a database of a medium-size German city with 130,000 listings. The system uses a new technique of taking a combined decision on the joint probability over multiple dialogue turns, and a dialogue strategy that strives to restrict the search space more and more with every dialogue turn. During the course of the dialogue, the last name of the desired subscriber must be spelled out. The spelling recognizer permits continuous spelling and uses a context-free grammar to parse common spelling expressions. This paper describes the system architecture, our maximum a-posteriori (MAP) decision rule, the spelling grammar, and the dialogue strategy. We give results on the SPEECHDAT and SIETILL databases on recognition of first names by spelling and on jointly deciding on the spelled and the spoken name. In a 35,000-names setup, the joint decision reduced name-recognition errors by 31%. 1.

Almost every speech application involves integration with real world databases which may be large or complex. Telephony based examples include call-centre automation, customer identification and directory assistance. Many such applications are intrinsically large vocabulary problems with complex data requirements.

Directory assistance systems are amongst the most challenging applications of speech recognition. Today, complete automation of the service fails because of the lacking accuracy of current speech recognizers, which are simply not able to differentiate between hundreds of thousands or even millions of different names occurring in large cities. In this paper, we show that this situation can be remedied by systematically combining all available knowledge sources (last names, first names, street names, partly including their spelled versions) in a statistically optimal way. Especially designed confidence measures for N-best lists are proposed to detect misrecognized turns.