Acquiring Information from Sources Responsive to Naturally-Spoken-Speech Commands Provided by a Voice-Enabled Device

Abstract

The present invention relates to a system for acquiring information from sources on a network, such as the Internet. A voice browsing system maintains a database containing a list of information sources, such as web sites, connected to a network. Each of the information sources is assigned a rank number which is listed in the database along with the record for the information source. In response to a speech command received from a user, a network interface system accesses the information source with the highest rank number in order to retrieve information requested by the user.

What is claimed: 
     
         1 . A system comprising:
 (a) at least one data processor, the at least one data processor operatively coupled to a plurality of communication data networks;   (b) at least one speaker-independent speech-recognition engine operatively coupled to the data processor;   (c) memory accessible to the at least one data processor and storing at least:
 (i) an instruction set for querying of information to be retrieved from a plurality of sources coupled to the plurality of communication data networks, the instruction set comprising:
 an indication of the plurality of sources, each identified by a source identifier, and each identifying certain information to be retrieved from the source identifier, and 
 
 (ii) at least one recognition grammar executable code corresponding to each instruction set and corresponding to data characterizing audio containing a naturally-spoken-speech command including an information request, 
   (d) wherein the at least one speaker-independent-speech-recognition engine is adapted to:
 (1) receive the data characterizing audio containing the naturally-spoken-speech command from a voice-enabled device via a first of the plurality of communication data networks; 
 (2) to recognize phenomes in the data characterizing audio containing naturally-spoken-speech commands to understand spoken words; and 
 (3) to generate recognition results data, 
   (e) wherein the at least one data processor is adapted to:
 (1) select the corresponding at least one recognition grammar executable code upon receiving the data characterizing the audio containing the naturally-spoken-speech command and to convert the data characterizing the audio containing the naturally-spoken-speech command into a data message for transmission to a network interface adapted to access a second of the plurality of communication data networks; and 
 (2) retrieve the instruction set corresponding to the recognition grammar executable codes provided by the at least one speaker-independent-speech-recognition engine and to access the information source queried by the instruction set to obtain at least a part of the information to be retrieved, and 
   (f) at least one speech-synthesis device operatively coupled to the at least one data processor, the at least one speech-synthesis device configured to produce an audio message relating to any resulting information retrieved from the plurality of information sources including a text-to-speech conversion of at least certain data in said any resulting information retrieved from the plurality of information sources, and to convey the audio message via the voice-enabled device.   
     
     
         2 . The system of  claim 1 , wherein the plurality of communication data networks includes the Internet. 
     
     
         3 . The system of  claim 1 , wherein the plurality of communication data networks include a local-area network. 
     
     
         4 . The system of  claim 1 , wherein the voice-enabled device is a home device. 
     
     
         5 . The method of  claim 1 , wherein the voice-enabled device is at least one of a group of an IP telephone, a cellular phone, a personal computer, a media player appliance, and a television or other video display device. 
     
     
         6 . The system of  claim 1 , wherein the speaker-independent-speech-recognition engine is adapted to analyze the phonemes to recognize conversational naturally-spoken-speech commands. 
     
     
         7 . The system of  claim 1 , wherein the speaker-independent-speech-recognition engine is adapted to recognize the naturally-spoken-speech commands. 
     
     
         8 . The system of  claim 1 , wherein the instruction set executable code further comprises: a content descriptor associated with each information-source identifier, the content descriptor pre-defining a portion of the information source containing the information to be retrieved. 
     
     
         9 . The system of  claim 1 , further comprising:
 a database operatively connected to the data processor, the database adapted to store the information gathered from the information sources in response to the information requests.   
     
     
         10 . The system of  claim 8 , wherein each recognition grammar executable code and each instruction set for querying of information to be retrieved are stored in the database. 
     
     
         11 . A method comprising:
 (a) providing at least one data processor, the data processor operatively coupled to a plurality of communication data networks;   (b) providing at least one speaker-independent-speech-recognition engine operatively coupled to the at least one data processor   (c) providing memory accessible to the data processor storing at least:
 (i) an instruction set for querying of the information to be retrieved from a plurality of sources coupled to the plurality of communication data networks, the instruction set comprising: an indication of the plurality of sources, each identified by an information-source identifier, and each identifying certain information to be retrieved from the information-source identifier, and 
 (ii) at least one recognition grammar executable code corresponding to each instruction set and corresponding to data characterizing audio containing a naturally-spoken-speech command including an information request, 
   (d) the at least one speaker-independent-speech-recognition engine:
 (i) receiving the data characterizing audio containing the naturally-spoken-speech command from the voice-enabled device via a first of the communication data networks, 
 (ii) recognizing phenomes in the data characterizing audio containing the naturally-spoken-speech commands to understand spoken words, and 
 (iii) generating recognition-results data, 
   (e) the least one data processor programmed to:
 (i) select the corresponding at least one recognition grammar executable code upon receiving the data characterizing audio containing the naturally-spoken-speech command and convert the data characterizing audio containing naturally-spoken-speech command into a data message for transmission to a network interface adapted to access a second of the one communication networks; and 
 (ii) retrieve the instruction set corresponding to the recognition grammar executable provided by the at least one speaker-independent-speech-recognition device and access the information source identified by the instruction set to obtain at least a part of the information to be retrieved; and 
   (f) providing at least one speech-synthesis device operatively connected to the at least one data processor, and by the at least one speech-synthesis device adapted to:
 (i) produce an audio message relating to any resulting information retrieved from the plurality of information sources including text-to-speech conversion of said any resulting information retrieved from the plurality of information sources, and 
 (ii) transmit the audio message via the voice-enabled device. 
   
     
     
         12 . The method of  claim 11 , wherein the plurality of communication data networks includes the Internet. 
     
     
         13 . The method of  claim 11 , wherein the plurality of communication data networks include a local-area network. 
     
     
         14 . The method of  claim 11 , wherein the voice-enabled device is a telephone. 
     
     
         15 . The method of  claim 11 , wherein the speaker-independent-speech-recognition engine is adapted to analyze the phonemes to recognize conversational naturally-spoken-speech commands. 
     
     
         16 . The method of  claim 11 , wherein the speaker-independent-speech-recognition engine is adapted to recognize the naturally-spoken-speech commands. 
     
     
         17 . The method of  claim 11 , wherein the instruction set executable code further comprises: a content descriptor associated with each information-source identifier, the content descriptor pre-defining a portion of the information source containing the information to be retrieved. 
     
     
         18 . The method of  claim 11 , further comprising:
 providing a database and operatively connecting the database to the data processor and storing the information gathered from the information sources in response to the information requests in the database.   
     
     
         19 . The method of  claim 18 , wherein each recognition grammar executable code and each instruction set are stored in the database. 
     
     
         20 . The method of  claim 18 , wherein the voice-enabled device is at least one of a group of an IP telephone, a cellular phone, a personal computer, a media player appliance, and a television or other video display device.