Learning new user formulations in automatic directory assistance

Abstract

Telecom Italia has deployed since the beginning of year 2001 a nationwide automatic Directory Assistance (DA) system that routinely serves customers asking for residential and business listings. DA for business listings is a challenging task: one of its main problems is that customers formulate their requests for the same listing with a great variability. Since it is difficult to reliably predict a priori the user formulations, in this paper we propose a procedure for detecting, from field data, user formulations that were not foreseen by the designers. These formulations can be added, as variants, to the denominations already included in the system to reduce its failures. We show that using a large database associating phonetic transcriptions of user utterances with the phone number provided by the automatic service, a completely unsupervised approach detects most of the old formulations. Furthermore, our procedure is able to filter a huge amount of calls routed to the operators, and to detect a limited number of phonetic strings that are good candidate to be included as new formulation variants in the system vocabulary.

Learning Pronunciation and Formulation Variants in Continuous Speech Applications

“…In the Directory Assistance application described in [3] the task of learning formulation variants has been addressed, for isolated words, by collecting and clustering the phonetic strings corresponding to user requests that the automatic system was not able to complete. The same approach has been extended in this work, to generic grammar directed tasks.…”

“…This paper presents the approaches that have been devised to detect vocabulary words pronunciation variants and linguistic formulations not covered by the recognition grammar. This work extends to grammars the phonetic learning approach that has been successfully applied, for isolated word recognition, to an automated Directory Assistance system [3]. This service, developed for Telecom Italy, is in operation since 2000, deals with both business and residential requests from a database of 25 million Italian subscribers.…”