System and method for generating manually designed and automatically optimized spoken dialog systems
View Patent ↗Disclosed herein are systems, computer-implemented methods, and tangible computer-readable storage media for generating a natural language spoken dialog system. The method includes nominating a set of allowed dialog actions and a set of contextual features at each turn in a dialog, and selecting an optimal action from the set of nominated allowed dialog actions using a machine learning algorithm. The method includes generating a response based on the selected optimal action at each turn in the dialog. The set of manually nominated allowed dialog actions can incorporate a set of business rules. Prompt wordings in the generated natural language spoken dialog system can be tailored to a current context while following the set of business rules. A compression label can represent at least one of the manually nominated allowed dialog actions.
1. A method comprising:
at each turn in a dialog, nominating via a processor, using a partially observable Markov decision process in parallel with a conventional dialog state, a set of allowed dialog actions and a set of contextual features; and
generating a response based on the set of contextual features and a dialog action selected, via a machine learning algorithm, from the set of allowed dialog actions.
2. The method of claim 1 , further comprising using reinforcement learning to augment the machine learning algorithm.
3. The method of claim 2 , wherein the machine learning algorithm is based on the partially observable Markov decision process.
4. The method of claim 1 , wherein the set of allowed dialog actions comprises a set of business rules.
5. The method of claim 4 , further comprising tailoring wordings in a natural language spoken dialog system based on a current context and the set of business rules.
6. The method of claim 1 , wherein a lower-dimensional feature vector represents one of the set of allowed dialog actions.
7. The method of claim 1 , further comprising receiving, during the dialog, disambiguating information from a user, wherein applying the disambiguating information removes dialog actions from the set of allowed dialog actions.
8. A system comprising:
a processor; and
a computer-readable storage medium having instructions stored which, when executed by the processor, cause the processor to perform operations comprising:
at each turn in a dialog nominating, using a partially observable Markov decision process in parallel with a conventional dialog state, a set of allowed dialog actions and a set of contextual features; and
generating a response based on the set of contextual features and a dialog action selected, via a machine learning algorithm, from the set of allowed dialog actions.
9. The system of claim 8 , the computer-readable storage medium having additional instructions stored which result in operations comprising using reinforcement learning to augment the machine learning algorithm.
10. The system of claim 9 , wherein the machine learning algorithm is based on the partially observable Markov decision process.
11. The system of claim 8 , wherein the set of allowed dialog actions comprises a set of business rules.
12. The system of claim 11 , the computer-readable storage medium having additional instructions stored which result in operations comprising tailoring wordings in a natural language spoken dialog system based on a current context and the set of business rules.
13. The system of claim 8 , wherein a lower-dimensional feature vector represents one of the set of allowed dialog actions.
14. The system of claim 8 , the computer-readable storage medium having additional instructions stored which result in operations comprising receiving, during the dialog, disambiguating information from a user, wherein applying the disambiguating information removes dialog actions from the set of allowed dialog actions.
15. A computer-readable storage device having instructions stored which, when executed by a computing device, cause the computing device to perform operations comprising:
at each turn in a dialog nominating, using a partially observable Markov decision process in parallel with a conventional dialog state, a set of allowed dialog actions and a set of contextual features; and
generating a response based on the set of contextual features and a dialog action selected, via a machine learning algorithm, from the set of allowed dialog actions.
16. The computer-readable storage device of claim 15 , the computer-readable storage medium having additional instructions stored which result in operations comprising using reinforcement learning to augment the machine learning algorithm.
17. The computer-readable storage device of claim 16 , wherein the machine learning algorithm is based on the partially observable Markov decision process.
18. The computer-readable storage device of claim 15 , wherein the set of allowed dialog actions comprises a set of business rules.
19. The computer-readable storage device of claim 18 , the computer-readable storage medium having additional instructions stored which result in operations comprising tailoring wordings in a natural language spoken dialog system based on a current context and the set of business rules.
20. The computer-readable storage device of claim 15 , wherein a lower-dimensional feature vector represents one of the set of allowed dialog actions.