Automatic dialog system with database language model
An automatic dialog system for spoken inquiries into a database entry which contains several components available for inquiry, wherein speech recognition of a spoken utterance for inquiring into the database entry is supported by a language model which was prepared before a start of the dialog to which the spoken utterance belongs and which models a relative frequency of correlated occurrence of the components of the database entry provided for the inquiry in the spoken utterance of the dialog.
1. An automatic dialog system for spoken inquiries into a database entry which contains components available for inquiry, wherein speech recognition of a spoken utterance for inquiring into the database entry is supported by a language model which is prepared before a start of a dialog to which the spoken utterance belongs and which models a relative frequency of occurrence of the components of the database entry provided for the inquiry in the spoken utterance of the dialog, wherein the relative frequency is independent of an ordering of words in the spoken utterance.
2. The automatic dialog system for spoken database inquiries of claim 1 ,
wherein the language model assigns a score 0 to the occurrence of components of a data record in the spoken utterance which is not contained in the inquired database.
3. The automatic dialog system for spoken database inquiries of claim 1 ,
wherein the language model assigns to the occurrence of components of a data record in the spoken utterance which is contained in the inquired database a score in accordance with their relative frequency in the inquired database, and
wherein the language model assigns the score 0 to the occurrence of components of a data record in the spoken utterance which is not contained in the inquired database.
4. The automatic dialog system for spoken database inquiries of claim 1 ,
wherein the language model assigns to the occurrence of components of a data record in the spoken utterance which is contained in the inquired database a score in accordance with their relative frequency in the inquired database, and
wherein the language model assigns to the occurrence of components of a data record in the spoken utterance which is not contained in the inquired database a score in accordance with the relative frequency of such spoken utterance.
5. The automatic dialog system for spoken database inquiries of claim 1 , wherein said components are distributed over several utterances.
6. A speech recognition system for recognizing a spoken utterance for the spoken inquiry after a database entry which comprises components available for inquiry, wherein the speech recognition system is supported by a language model which is prepared before a start of a dialog to which the spoken utterance belongs and which models a relative frequency of occurrence of the components of the database entry provided for the inquiry in the spoken utterance of the dialog, wherein the relative frequency is independent of an ordering of words in the spoken utterance.
7. The speech recognition system of claim 6 , wherein said components are distributed over several utterances.
8. A computer readable medium encoded with a data structure comprising a language model for supporting speech recognition of a spoken utterance for spoken inquiries of a database entry which comprises components available for inquiry,
wherein the language model is prepared before a start of the dialog to which the spoken utterance belongs, and wherein the language model models a relative frequency of occurrence of the components of the database entry provided for the inquiry in the spoken utterance of a dialog, wherein the relative frequency is independent of an ordering of words in the spoken utterance.
9. The computer readable medium of claim 8 , wherein said components are distributed over several utterances.
10. A method of orally inquiring into a database entry which comprises components available for inquiry, wherein the speech recognition of a spoken utterance for inquiring into the database entry is supported by a language model which was constructed before the start of a dialog to which the spoken utterance belongs and which models the relative frequency of occurrence of the component of the database entry provided for the inquiry in the spoken utterance of the dialog, wherein the relative frequency is independent of an ordering of words in the spoken utterance.
11. The method of claim 10 , wherein said components are distributed over several utterances.