Name dialing using final user defined vocabularies in mobile (GSM and TACS) and fixed telephone networks

This paper presents the results obtained on the evaluation of a new approach for generation of phonetic transcriptions for name dialing applications in different telephone networks and with temporal variations. In this kind of applications on-line construction of user vocabularies is mandatory. The proposed method allows adaptive selection of new transcriptions requiring much less speech utterances for system training than other approaches. The new approach is evaluated using data from different telephone networks (PSTN, GSM and TACS networks) and from different temporal utterances (recordings done in a period of two months)