Speech dialog method and system
View Patent ↗An electronic device ( 300 ) for speech dialog includes functions that receive ( 305, 105 ) a speech phrase that comprises a request phrase that includes an instantiated variable ( 215 ), generate ( 335, 115 ) pitch and voicing characteristics ( 315 ) of the instantiated variable, and performs speech recognition ( 319, 125 ) of the instantiated variable to determine a most likely set of acoustic states ( 235 ). The electronic device may generate ( 335, 140 ) a synthesized value of the instantiated variable using the most likely set of acoustic states and the pitch and voicing characteristics of the instantiated variable. The electronic device may use a table of previously entered values of variables that have been determined to be unique, and in which the values are associated with a most likely set of acoustic states and the pitch and voicing characteristics determined at the receipt of each value to disambiguate ( 425, 430 ) a newly received instantiated variable.
1. A method for speech dialog, comprising:
receiving a speech phrase that comprises a request phrase that includes an instantiated variable;
generating pitch and voicing characteristics of the instantiated variable;
performing speech recognition of the instantiated variable to determine a most likely set of acoustic states; and
generating a synthesized value of the instantiated variable using the most likely set of acoustic states and the pitch and voicing characteristics of the instantiated variable.
2. A method for speech dialog according to claim 1 , wherein the request phrase further includes a non-variable segment that is associated with the instantiated variable, further comprising:
performing speech recognition of the non-variable segment; and
presenting an acoustically stored response phrase.
3. The method for speech dialog according to claim 1 , wherein the performing of the speech recognition of the instantiated variable comprises:
determining acoustic characteristics of the instantiated variable; and
using a mathematical model of stored lookup values and the acoustic characteristics to determine the most likely set of acoustic states.
4. The method for speech dialog according to claim 3 , wherein the mathematical model of stored lookup values is a hidden Markov model.
5. The method for speech dialog according to claim 1 , wherein generating the synthesized value of the instantiated variable is performed when a metric of the most likely set of acoustic states meets a criterion, and further comprising:
presenting an acoustically stored out-of-vocabulary response phrase when the metric of the most likely set of acoustic states fails to meet the criterion.
6. An electronic device for speech dialog, comprising:
means for receiving a speech phrase that comprises a request phrase that includes an instantiated variable;
means for generating pitch and voicing characteristics of the instantiated variable;
means for performing speech recognition of the instantiated variable to determine a most likely set of acoustic states; and
means for generating a synthesized value of the instantiated variable using the most likely set of acoustic states and the pitch and voicing characteristics of the instantiated variable.
7. The electronic device for speech dialog according to claim 6 , wherein the request phrase further includes a non-variable segment that is associated with the instantiated variable, further comprising:
means for performing speech recognition of the non-variable segment; and
means for presenting an acoustically stored response phrase.
8. The electronic device for speech dialog according to claim 6 , wherein the performing of the speech recognition of the instantiated variable comprises:
means for determining acoustic characteristics of the instantiated variable; and
means for using a stored model of acoustic states and the acoustic characteristics to determine the most likely set of acoustic states.
9. The electronic device for speech dialog according to claim 6 , wherein generating the synthesized value of the instantiated variable is performed when a metric of the most likely set of acoustic states meets a criterion, and further comprising:
means for presenting an acoustically stored out-of-vocabulary response phrase when the metric of the most likely set of acoustic states fails to meet the criterion.
10. A media that includes a stored set of program instructions, comprising:
a function for receiving a speech phrase that comprises a request phrase that includes an instantiated variable;
a function for generating pitch and voicing characteristics of the instantiated variable;
a function for performing speech recognition of the instantiated variable to determine a most likely set of acoustic states; and
a function for generating a synthesized value of the instantiated variable using the most likely set of acoustic states and the pitch and voicing characteristics of the instantiated variable.
11. The media according to claim 10 , wherein the request phrase further includes a non-variable segment that is associated with the instantiated variable, further comprising:
a function for performing speech recognition of the non-variable segment; and
a function for presenting an acoustically stored response phrase.
12. The media according to claim 10 , wherein the function for performing the speech recognition of the instantiated variable comprises:
a function for determining acoustic characteristics of the instantiated variable; and
a function for using a mathematical model of stored lookup values and the acoustic characteristics to determine the most likely set of acoustic states.
13. The method for speech dialog according to claim 12 , wherein the mathematical model of stored lookup values is a hidden Markov model.
14. The media according to claim 10 , wherein the function of generating the synthesized value of the instantiated variable is performed when a metric of the most likely set of acoustic states meets a criterion, and further comprising:
a function for presenting an acoustically stored out-of-vocabulary response phrase when the metric of the most likely set of acoustic states fails to meet the criterion.
15. A method for speech dialog, comprising:
receiving one or more speech phrases that include one or more instantiated variables;
generating pitch and voicing characteristics of the one or more instantiated variables;
performing speech recognition of the instantiated variables to determine a most likely set of acoustic states for each of the one or more instantiated variables;
storing most likely sets of acoustic states and pitch and voicing characteristics of the one or more instantiated variables that are each determined to correctly correspond to a unique value of a variable into a value attributes database, each of the most likely sets of acoustic states and pitch and voicing characteristics being stored in association with the unique value to which it corresponds;
receiving a new speech phrase that comprises a new request phrase that includes a new instantiated variable;
determining by speech recognition of the new instantiated variable a recognized value;
determining a most likely set of acoustic states and pitch and voicing characteristics from the value attributes database that has a likelihood of disambiguating the recognized value;
generating a response using the most likely set of acoustic states and the pitch and voicing characteristics associated with the recognized value in the value attributes database.