A Textual processor to handle ATIS queries

A Textual processor to handle ATIS queries Douglas O'Shaughnessy INRS-Telecommunications University of Quebec 3 Place du Commerce Nuns Island, Quebec, Canada H3E 1H6 ABSTRACT This paper describes the initial development of a natural language text processor, as the first step in an INRS dialogue-by-voice system. The eventual system will accept natural, spontaneous speech from users and produce responses from the databases in the form of synthetic speech. This paper reports results in processing the textual version of ATIS (Air Travel Information System) queries. The current system (programmed in C) accepts as input the cleaned-up text (SNOR) ver- sion of the spoken queries, and produces the desired Official Airline Guide (OAG) information as output. It uses only the words in the in- put text, and not any punctuation marks, on the assumption that such marks are difficult to obtain directly from speech input. Based on the training text data, the system correctly interprets a large majority of the textual queries. INTRODUCTION Speech recognition systems have made significant progress in recent years toward the goal of correctly interpreting continuously-spoken utterances. However, substantial restric- tions are usually imposed upon the speaker, to guarantee success. Typically, one must either pause after each word, restrict one's choice of words to a small vocabulary, and/or train the system to adapt to one's voice. In many systems, it is not feasible to insist on vocabulary restrictions, nor can the system always be trained ahead of time to a user's voice. Many applications over the telephone to serve the general public will be of this latter type~ Furthermore, most users do not like altering their speak- ing style, and especially not speaking in isolated-word formant. Thus most practical applications of the future will have to be speaker-independent (i.e., trained ahead of time by other speak- ers), without major restrictions in vocabulary, and be able to accept normal, spontaneous speech. In particular, one major application is allowing the general public to do transactions directly with computer databases (in- cluding over the telephone). As an example of this type of interac- tion, we are currently examining a system to permit direct access for a user to air travel information. A user can pose natural ques- tions to the database and receive answers just as a travel agent does. The database is that of the Official Airline Guide (OAG). To simplify the task slightly, we use a subset of the flights in the OAG: only those for airports at nine major US cities (Atlanta, Boston, Baltimore, Denver, Dallas, Oakland, Philadelphia, Pitts- burgh, and San Francisco). Otherwise, the entire OAG database is used. In the future, we will investigate actual voice dialogues be- tween a user and the database, but for now the subject of this study is limited to the analysis of individual queries by users. We wish to design an automatic system to correctly respond to the user with the desired OAG information. As a first step toward this goal, the current study is further limited to the analysis of …

Content truncated to 500 words