Information communication terminal, information communication system, information communication method, and storage medium for storing an information communication program thereof for recognizing speech information

• US 8,126,712 B2
• Filed: 02/08/2006
• Issued: 02/28/2012
• Est. Priority Date: 02/08/2005
• Status: Expired due to Fees

• Current Assignee: Nippon Telegraph and Telephone Corporation
• Original Assignee: Nippon Telegraph and Telephone Corporation
• Inventors: Mukaigaito, Takeya, Takada, Shinya, Yokozeki, Daigoro, Sakai, Miki, Sakai, Rie, Arai, Katsuya, Nishihara, Takuo, Murayama, Takahiko
• Primary Examiner(s): Godbold, Douglas
• Application Number:
  US11/815,839
• Publication Number:
  US 20090018832A1
• Time in Patent Office: 2,211 Days
• Field of Search: 704/251, 704/270, 704/231, 704/242, 704/255
• US Class Current: 704/251
• CPC Class Codes: G10L 15/1815 Semantic context, e.g. disambiguation of the recognition hypotheses based on word meaning
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: H04M 1/2535 adapted for voice communication over an Internet Protocol [IP] network Voice over Internet Protocol (VoIP) network equipment and services H04M7/006; implementation or adaptation of Internet protocol [IP], of transmission control protocol [TCP] or of user datagram protocol [UDP] H04L69/16
• CPC Class Codes: H04M 2201/38 Displays
• CPC Class Codes: H04M 2201/40 using speech recognition
• CPC Class Codes: H04M 2250/64 file transfer between terminals
• CPC Class Codes: H04M 2250/74 with voice recognition means
• CPC Class Codes: H04M 3/4938 comprising a voice browser which renders and interprets, e.g. VoiceXML

Abstract

An information communication terminal (100) that includes: a speech recognition module (6) for recognizing speech information to identify a plurality of words in the recognized speech information; a storage medium (20) for storing keyword extraction condition setting data (24) in which a condition for extracting a keyword is set; a keyword extraction module (8) for reading the keyword extraction condition setting data (24) to extract a plurality of keywords from the plurality of words; a related information acquisition module (11) for acquiring related information related to a plurality of keywords; and a related information output module (14) for providing related information to a monitor (2).

Claims

1. An information communication terminal configured to exchange at least speech information with a plurality of information communication terminals, comprising:

a speech recognition module configured to recognize the speech information to identify a plurality of words based on the recognized speech information;

a storage medium configured to store keyword extraction condition setting data in which conditions for extracting keywords are set;

a keyword extraction module configured to read the keyword extraction condition setting data to extract a plurality of keywords from the plurality of words;

a subject extraction processing module configured to;

associate the plurality of words read by the keyword extraction module with knowledge network data in which the plurality of keywords and a route among the plurality of keywords are described in a network form,

generate a plurality of word pairs in a predetermined order from the plurality of keywords,

extract the shortest route connecting words in each word pair from the knowledge network data,

give point values to each word on each of the shortest routes,

count the point values given to the respective words, and

extract a word having a relatively high point value as a subject word;

a related information acquisition module configured to acquire related information related to the plurality of keywords; and

a related information output module configured to provide the related information to a monitor.

2. The information communication terminal according to claim 1, further comprising:

a keyword output module configured to provide the plurality of keywords to the monitor; and

an information acquisition condition acceptance module configured to accept keywords selected from the plurality of keywords and an input of an information acquisition condition of the related information,

wherein the related information acquisition module reads the accepted keyword and the accepted information acquisition condition to acquire the related information.

3. The information communication terminal according to claim 2, further comprising:

a common information module configured to share at least one of the information acquisition condition and the related information with another information communication terminal.

4. The information communication terminal according to claim 1, further comprising:

a speech recognition result exchange module configured to exchange a speech recognition result including the plurality of words specified based on the speech information with another information communication terminal.

5. The information communication terminal according to claim 1, further comprising:

an emotion recognition module configured to recognize emotion information acquired from the speech information;

an emotion information transmission module configured to transmit the emotion information to another information communication terminal; and

an emotion information reception module configured to receive emotion information sent from the another information communication terminal.

6. The information communication terminal according to claim 1, further comprising:

a warning display module configured to compare a warning object keyword set stored in the storage medium with the plurality of keywords to calculate an appearance number at which a keyword included in the warning object keyword set appears to display a warning when the appearance number is equal to or higher than a previously-set number.

7. The information communication terminal according to claim 1, further comprising:

a communication cutoff module configured to compare a warning object keyword set stored in the storage medium with the plurality of keywords to calculate an appearance number at which a keyword included in the warning object keyword set appears to cut the communication with the plurality of information communication terminals when the appearance number is equal to or higher than a previously set number.

8. The information communication terminal according to claim 1, further comprising:

a speech information recording module configured to compare a warning object keyword set stored in the storage medium with the plurality of keywords to calculate an appearance number at which a keyword included in the warning object keyword set appears to start the recording of the speech information when the appearance number is equal or higher than a previously set number.

9. The information communication terminal according to claim 1, wherein the keyword extraction module further comprises:

a different meaning word meaning determination module, including;

a different meaning word determination section configured to determine whether or not the plurality of words include a homonym or a homographe, by using knowledge network data in which vocabulary information showing a correspondence between a vocabulary and a reading sound of the vocabulary and link information showing a distance relation of the vocabulary; and

a meaning determination section configured to determine the meaning of the word having a different meaning, based on the distance between the plurality of words.

10. The information communication terminal according to claim 9, wherein the different meaning word determination section further includes:

a knowledge network data check section configured to refer to the vocabulary information and the link information;

a different meaning word existence check section configured to refer to a result of checking the knowledge network data check section to determine whether a homonym or a homographe exists or not; and

a vocabulary acquisition section configured to refer, when a homonym or a homographe exists, to the vocabulary information to acquire all vocabulary having an identical notation or reading sound with that of the word determined as a homonym or a homographe.

11. The information communication terminal according to claim 10, wherein the meaning determination section further includes:

a distance determination section configured to compare the respective pieces of vocabulary acquired by the vocabulary acquisition section with vocabulary in a word collection other than a homonym or a homographe to investigate whether a distance between pieces of vocabulary is within a predetermined distance or not;

a distance count section configured to count a distance between pieces of vocabulary for the respective pieces of vocabulary; and

a maximum count number value determination section configured to determine vocabulary having the maximum count number as vocabulary of a homonym or a homographe.

12. The information communication terminal according to claim 1, wherein the subject extraction processing module gives, when the words have therebetween a plurality of the shortest routes, points to words existing on the plurality of the shortest routes.

13. The information communication terminal according to claim 1, wherein the subject extraction processing module compares the shortest route with a previously set threshold value to give the point only to a word existing on the shortest route having a distance shorter than the threshold value.

14. The information communication terminal according to claim 1, wherein the subject extraction processing module gives higher points to a word existing on the shortest route that has a shorter distance to a word positioned at an end of the shortest route.

15. The information communication terminal according to claim 14, wherein the subject extraction processing module reads a previously set quantity to extract, from among words existing on the shortest route, words having the higher points in the quantity.

16. The information communication terminal according to claim 1, wherein the subject extraction processing module reads a previously set threshold value to extract a word having the point equal to or higher than a threshold value.

17. The information communication terminal according to claim 1, wherein the subject extraction processing module calculates, depending on a distance of the shortest route, an evaluation value of a word positioned at an end of the shortest route to give the word existing on the shortest distance a point considering the evaluation value.

18. The information communication terminal according to claim 17, wherein the subject extraction processing module gives a higher point to a word existing on the shortest distance as the shortest route has a shorter distance.

19. The information communication terminal according to claim 17, wherein the subject extraction processing module gives a higher point to a word of words existing on the shortest route that has a shorter distance to the word positioned at an end of the shortest route.

20. The information communication terminal according to claim 1, wherein the keyword extraction module uses knowledge network data in which data representing a semantic relation between two words by a distance are prepared for a plurality of words to calculate data of the shortest distance between extracted words to calculate, while excluding the shortest distance data equal to or higher than a fixed distance, a relation between extracted words to extract the keyword based on the calculation result of the relation.

21. The information communication terminal according to claim 20, wherein the keyword extraction module accumulates, with regards to the respective extracted words, appearance numbers of the shortest distance to another extracted word for each the shortest distance to calculate the relation between the words to extract, as the keyword, a word for which an accumulation value of the shortest distance is equal to or higher than a set value.

22. The information communication terminal according to claim 20, wherein the keyword extraction module calculates, with regards to the respective extracted words, an average distance of the shortest distance data to calculate a relation between the words to extract, as the keyword, a word for which the average distance is equal to or higher than a set value.

23. The information communication terminal according to claim 20, wherein the keyword extraction module calculates, with regards to the respective extracted words, a difference between an average distance between words shown by the knowledge network data and an average distance of the shortest distance data to calculate a relation between words to extract, as the keyword, a word for which the difference is equal to or higher than a set value.

24. The information communication terminal according to claim 23, wherein the keyword extraction module reads the keyword to refer to the knowledge network data to extract a related word for having a distance from the keyword equal to or lower than a set value.

25. The information communication terminal according to claim 23, wherein the related word extraction module reads an upper limit value of an extraction quantity of the related word to extract the related words in a quantity corresponding to the upper limit value.

26. An information communication system connected with a plurality of information communication terminals via a network, one of the information communication terminals includes:

a speech recognition module configured to recognize at least speech information sent via the network to identify a plurality of words, based on the recognized speech information;

a storage medium configured to store keyword extraction condition setting data in which conditions for extracting keywords are set;

a keyword extraction module configured to read the keyword extraction condition setting data to extract a plurality of keywords from the plurality of words;

a related information acquisition module configured to acquire related information related to the plurality of keywords; and

a related information output module configured to provide related information to a monitor,

wherein the keyword extraction module includes;

a subject extraction processing module configured to;

associate the plurality of words extracted from the speech information sent via the network with knowledge network data in which the plurality of keywords and a route among the plurality of keywords are described in a network form,

generate a plurality of word pairs in a predetermined order from the plurality of keywords,

extract the shortest route connecting words in each word pair from the knowledge network data,

give point values to words on each of the shortest routes,

count the point values given to the respective words, and

extract a word having a relatively high point value as a subject word.

27. The information communication system according to claim 26, wherein the information terminal further includes:

a warning information acquisition module configured to read operation condition setting data stored in the storage medium to acquire warning information for displaying a warning via the network, when the plurality of keywords extracted by the keyword extraction module include a specific keyword.

28. The information communication system according to claim 26, wherein the information terminal further includes:

a keyword extraction request means configured to read operation setting condition data stored in the storage medium and the keyword extraction condition setting data to request a keyword extraction server connected via the network to extract the keyword.

29. The information communication system according to claim 26, the keyword extraction means further comprises:

a different meaning word meaning determination module, including;

a different meaning word determination section configured to determine whether or not the plurality of words include a homonym or a homographe, by using knowledge network data in which vocabulary information showing a correspondence between a vocabulary and a reading sound of the vocabulary and link information showing a distance relation of the vocabulary; and

a meaning determination section configured to determine the meaning of the word having a different meaning, based on the distance between the plurality of words.

30. The information communication system according to claim 26, wherein the keyword extraction means uses the knowledge network data to calculate the shortest distance data between the plurality of extracted words to calculate, while excluding the shortest distance data equal to or higher than a fixed distance, a relation between extracted words to extract the keyword based from the calculation result of the relation.

31. An information communication method for using an information communication terminal including a speech recognition module, a storage medium, a keyword extraction module, a subject extraction module, a related information acquisition module, a related information output module, and a monitor to exchange at least speech information with another information communication terminal, the method comprising:

recognizing a plurality of words in the speech information by the speech recognition module;

storing keyword extraction condition setting data including conditions for extracting keywords in the storage medium;

reading the keyword extraction condition setting data from the storage medium and extracting a plurality of keywords from the plurality of words by the keyword extraction module;

associating the plurality of words read by the keyword extraction module with knowledge network data in which the plurality of keywords and a route among the plurality of keywords are described in a network form;

generating a plurality of word pairs in a predetermined order from the plurality of keywords;

extracting the shortest route connecting words in each word pair from the knowledge network data;

giving point values to words on each of the shortest routes;

counting the point values given to the respective words;

extracting a word having a relatively high point value as a subject word;

acquiring related information relating to the plurality of keywords by the related information acquisition module; and

providing the related information to the monitor by the related information output module.

32. A non-transitory computer-readable storage medium storing an information communication program for causing an information communication terminal including a speech recognition module, a storage medium, a keyword extraction module, a subject extraction module, a related information acquisition module, a related information output module, and a monitor to exchange at least speech information with another information communication terminal, to execute an information communication method comprising:

recognizing a plurality of words in speech information by the speech recognition module;

storing keyword extraction condition setting data including conditions for extracting keywords in the storage medium;

reading the keyword extraction condition setting data from the storage medium and extract a plurality of keywords from the plurality of words by the keyword extraction module;

associating the plurality of words read by the keyword extraction module with knowledge network data in which the plurality of keywords and a route among the plurality of keywords are described in a network form;

generating a plurality of word pairs in a predetermined order from the plurality of keywords;

extracting the shortest route connecting words in each word pair from the knowledge network data;

giving point values to words on each of the shortest routes;

counting the point values given to the respective words;

extracting a word having a relatively high point value as a subject word;

acquiring related information related to the plurality of keywords by the related information acquisition module; and

providing the related information by the related information output module.