System and method for generating manually designed and automatically optimized spoken dialog systems

Abstract

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.

We claim: 
     
         1 . A method comprising:
 nominating, via a processor configured to use a partially observable Markov decision process in parallel with a conventional dialog state, a set of contextual features; and   generating an audible response in a dialog between a user and a spoken dialog system based at least in part on the set of contextual features.   
     
     
         2 . The method of  claim 1 , further comprising:
 nominating, via the processor configured to use the partially observable Markov decision process in parallel with the conventional dialog state, a set of dialog actions; and   generating the audible response based on the set of dialog actions.   
     
     
         3 . The method of  claim 2 , further comprising generating the audible response based on the set of dialog actions and via a machine learning algorithm. 
     
     
         4 . The method of  claim 3 , further comprising augmenting the machine learning algorithm using reinforcement learning. 
     
     
         5 . The method of  claim 4 , wherein the machine learning algorithm augmented by the reinforcement learning is based on the partially observable Markov decision process. 
     
     
         6 . The method of  claim 3 , further comprising:
 assigning a reward to the set of dialog actions as part of the machine learning algorithm.   
     
     
         7 . The method of  claim 1 , further comprising tailoring wordings in a spoken dialog system associated with the dialog based on a current context and a set of business rules. 
     
     
         8 . A system comprising:
 a processor configured to use a partially observable Markov decision process in parallel with a conventional dialog state; and   a computer-readable storage medium having instructions stored which, when executed by the processor, cause the processor to perform operations comprising:
 nominating a set of contextual features; and 
 generating an audible response in a dialog between a user and a spoken dialog system based at least in part on the set of contextual features. 
   
     
     
         9 . The system of  claim 8 , further comprising:
 nominating, via the processor configured to use the partially observable Markov decision process in parallel with the conventional dialog state, a set of dialog actions; and   generating the audible response based on the set of dialog actions.   
     
     
         10 . The system of  claim 9 , wherein the computer-readable storage medium stores additional instructions stored which, when executed by the processor, cause the processor to perform operations further comprising:
 generating the audible response based on the set of dialog actions and via a machine learning algorithm.   
     
     
         11 . The system of  claim 10 , wherein the computer-readable storage medium stores additional instructions stored which, when executed by the processor, cause the processor to perform operations further comprising:
 augmenting the machine learning algorithm using reinforcement learning.   
     
     
         12 . The system of  claim 11 , wherein the machine learning algorithm augmented by the reinforcement learning is based on the partially observable Markov decision process. 
     
     
         13 . The system of  claim 10 , wherein the computer-readable storage medium stores additional instructions stored which, when executed by the processor, cause the processor to perform operations further comprising:
 assigning a reward to the set of dialog actions as part of the machine learning algorithm.   
     
     
         14 . The system of  claim 8 , wherein the computer-readable storage medium stores additional instructions stored which, when executed by the processor, cause the processor to perform operations further comprising:
 tailoring wordings in a spoken dialog system associated with the dialog based on a current context and a set of business rules.   
     
     
         15 . A computer-readable storage device having instructions stored which, when executed by a processor configured to use a partially observable Markov decision process in parallel with a conventional dialog state, cause the processor to perform operations comprising:
 nominating a set of contextual features; and   generating an audible response in a dialog between a user and a spoken dialog system based at least in part on the set of contextual features.   
     
     
         16 . The computer-readable storage device of  claim 15 , wherein the computer-readable storage device stores additional instructions stored which, when executed by the processor, cause the processor to perform operations further comprising:
 nominating, via the processor configured to use the partially observable Markov decision process in parallel with the conventional dialog state, a set of dialog actions; and   generating the audible response based on the set of dialog actions.   
     
     
         17 . The computer-readable storage device of  claim 16 , wherein the computer-readable storage device stores additional instructions stored which, when executed by the processor, cause the processor to perform operations further comprising:
 generating the audible response based on the set of dialog actions and via a machine learning algorithm.   
     
     
         18 . The computer-readable storage device of  claim 17 , further comprising augmenting the machine learning algorithm using reinforcement learning. 
     
     
         19 . The computer-readable storage device of  claim 18 , wherein the machine learning algorithm augmented by the reinforcement learning is based on the partially observable Markov decision process. 
     
     
         20 . The computer-readable storage device of  claim 17 , further comprising:
 assigning a reward to the set of dialog actions as part of the machine learning algorithm.