Systems and methods for interactively accessing hosted services using voice communications

Abstract

Systems and methods for an interactive voice response system are described herein. In one embodiment, the system may include a voice recognition module, a session manager, and a voice generator module. An utterance received at the voice recognition module may be converted into one or more structures using a lexicon tied to an ontology. Concepts in the utterance may then be identified. If sufficient information has been provided to identify a relevant service, corresponding text responses associated with that service may then be converted into voice messages by the voice generator.

What is claimed is: 
     
         1 . A method implemented using a device comprises a voice recognition module, a session manager, and a voice generator, the method comprising:
 Receiving, from a user, an utterance via the voice recognition module;   converting data representing the utterance into one or more structures using lexicon tied to an ontology;   identifying one or more concepts based on the structures;   selecting automatically by the session manager a service associated with the concepts; and   providing, based on the selected service, a voice message using the voice generator.   
     
     
         2 . The method of  claim 1 , further comprising converting the utterance into text data, wherein the data representing the utterance is the text data. 
     
     
         3 . The method of  claim 1 , further comprising:
 determining that the utterance does not include sufficient information;   generating a text message designed to elicit further information;   converting the text message to a voice message using the voice generator; and   communicating the voice message.   
     
     
         4 . The method of  claim 1 , further comprising, before providing the voice message, generating a text message based on the selected service; and
 converting the text message to the voice message.   
     
     
         5 . The method of  claim 1 , further comprising:
 receiving further utterances via the voice recognition module in response to the voice message communicated to the user;   converting the utterances into one or more structures using lexicon tied to an ontology;   identifying concepts in the utterances using the structures;   obtaining information to provide to the user based on the identified concepts;   generating another text message based on at least one of the obtained information or the identified concepts;   converting the text message to a voice message using the voice generator; and   communicating the text message to the user.   
     
     
         6 . The method of  claim 1 , wherein converting the data representing the utterance into one or more structures using lexicon tied to ontology comprises associating the utterance with one or more units linked to a plurality of concepts, which in turn are associated with a plurality of services. 
     
     
         7 . The method of  claim 1 , wherein identifying the one or more concepts based on the structures comprises determining which concepts are linked with the units. 
     
     
         8 . The method of  claim 1 , further comprising, before the selecting, determining which service is most closely related to the utterance based on a number of associated concepts selected based on at least a portion of the utterance. 
     
     
         9 . The method of  claim 1 , further compromising identifying special concepts associated the selected service and assigning a value to the identified special concepts based on information included in the utterance. 
     
     
         10 . A system comprising for providing a service to a user comprising:
 a voice recognition module configured to receive an utterance and convert the utterance into text; one or more structures using lexicon tied to an ontology;   a session manger configured coupled with the voice recognition module and configured to:
 generate a structure associated with the text using lexicon tied to an ontology, identify concepts in the utterance using the structure; 
 provided the utterance includes sufficient information, select a service based on the concepts, and 
 generating a reply text message based on the selected service; and 
   a voice generator module coupled with the session manager and configured to convert the reply text message to a voice message.   
     
     
         11 . The system of  claim 10 , wherein the session manger is further configured, when the utterance does not include sufficient information, to generate a text message designed to elicit further information, and wherein the voice generator module is further configured to convert the text message to a voice message, and communicating the voice message to the user. 
     
     
         12 . The system of  claim 10 , wherein the session manager is further configured to generate a structure associated with the text by associating the utterance with one or more units linked to a plurality of concepts, which in turn are associated with a plurality of services. 
     
     
         13 . The system of  claim 12 , wherein the session manger is further configured to identify concepts in the utterance using the structures by determining which concepts are linked with the units. 
     
     
         14 . The system of  claim 13 , the session manger is further configured to select a service based on the concepts by determining which service is most closely related to the utterance based on the number of associated concepts that are linked to the utterance via the units for each of the plurality of services. 
     
     
         15 . The system of  claim 10 , wherein the session manger is further configured to identify special concepts associated the selected service and assign a value to the identified special concepts based on information included in the utterance. 
     
     
         16 . A non-transitory computer readable medium having stored therein computer executable instructions for:
 receiving, from a user, an utterance via a voice recognition module;   converting data representing the utterance into one or more structures using lexicon tied to an ontology;   identifying one or more concepts based on the structures;   selecting automatically by a session manager a service associated with the concepts; and   providing, based on the selected service, a voice message using a voice generator.   
     
     
         17 . The method of  claim 16 , further comprising:
 determining that the utterance does not include sufficient information;   generating a text message designed to elicit further information;   converting the text message to a voice message using the voice generator; and   communicating the voice message.   
     
     
         18 . The method of  claim 16 , further comprising, before providing the voice message, generating a text message based on the selected service; and
 converting the text message to the voice message.   
     
     
         19 . The method of  claim 16 , wherein identifying the one or more concepts based on the structures comprises determining which concepts are linked with the units. 
     
     
         20 . The method of  claim 16 , further comprising, before the selecting, determining which service is most closely related to the utterance based on a number of associated concepts selected based on at least a portion of the utterance.