JP2015099290A - In-utterance important word extraction device and in-utterance important word extraction using the device, and method and program thereof - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an in-utterance important word extraction device capable of extracting an important word from utterance information without creating a keyword list for each user.SOLUTION: A morpheme analysis part outputs a morpheme string comprising information on a word and a part of speech divided into morphemes through morpheme analysis of text information input to a text input part. A word registration part receives the morpheme string as an input and registers a word meeting a predetermined condition as a keyword in a keyword list. A speech recognition part creates and outputs a text sentence comprising a word string divided into morphemes through speech recognition processing on speech information input to a speech input part. An important word extraction part receives the text sentence as an input and extracts, as an important word, a word matching any of keywords registered in the keyword list.

Description

  The present invention relates to an important word extraction device in an utterance for extracting an important word in an utterance, an important word extraction system in an utterance using the device, a method and a program thereof.

  We communicate with others by speaking in our lives. The utterance includes various utterance information. By efficiently extracting important words from the utterance information, it is possible to create personal activity histories such as life logs and memorandums. it is conceivable that.

  Considering the convenience of the user, in order to leave a history, instead of uttering each time, the voice is always recorded, information can be obtained from the utterance information uttered in the usual unconscious life Is preferred. Personal activity histories and memorandums using words considered important obtained from the utterance information are considered to be useful records.

  Conventionally, as a method for extracting an important word from utterance contents, a method is known in which a recognition result matching a keyword is extracted as an important word using a keyword list in which the important word is registered in advance (for example, Patent Document 1).

JP 2008-286922 A

  However, since words that should be important words are considered to be different for each user, the conventional method has a problem that a keyword list must be created for each user. In addition, since it is generally considered that the object of interest of each user changes in daily life, it is necessary to perform daily maintenance of keywords directly connected to the object of interest. However, daily maintenance of the keyword list is not practical. In addition, since the speech rate of conversational speech in a normal conversation is fast and often produces unclear pronunciation, a misrecognized word may be erroneously extracted as an important word.

  The present invention has been made in view of such problems, and it is possible to automatically update a keyword list and reduce the risk of extracting misrecognized words as important words, It is an object of the present invention to provide a system for extracting important words in an utterance using the apparatus, a method and a program thereof.

  The important word extraction device in an utterance of the present invention includes a keyword list, a text input unit, a morpheme analysis unit, a word registration unit, a voice input unit, a voice recognition unit, and an important word extraction unit. Keywords are registered in the keyword list. Text information is input to the text input unit. The morpheme analysis unit outputs a morpheme string composed of word and part of speech information obtained by dividing the text information input to the text input unit into a morpheme. The word registration unit inputs the morpheme string output by the morpheme analysis unit and registers words satisfying a predetermined condition in the keyword list as keywords. Voice information is input to the voice input unit. The voice recognition unit generates and outputs a text sentence composed of a word string obtained by dividing the voice information input to the voice input unit by voice recognition processing into morphemes. The important word extraction unit receives a text sentence output from the speech recognition unit as an input, and extracts a word that matches one of the keywords registered in the keyword list as an important word.

  Moreover, the important word extraction system in an utterance of the present invention includes a voice input terminal, an important word extraction device in an utterance, a network, and a recognized word extraction server. The voice input terminal outputs the voice signal picked up by the microphone to the important word extraction device in the utterance. In-speech important word extraction device registers a word satisfying a predetermined condition obtained from input text information as a keyword, and a person utters from a voice file in which a voice signal output from a voice input terminal is recorded as a voice file The speech signal of the speech segment obtained by extracting only the speech part and the utterance start time information are transmitted to the recognition word extraction server via the network, and the word of the recognition word information that matches the keyword received from the recognition word extraction server is important. Extract and output as words. The recognized word extraction server receives a speech signal and speech start time information from the speech important word extraction device, and performs a speech recognition process on the speech signal in the speech section and converts a text sentence including a word string divided into morphemes. It generates and outputs the words constituting the text sentence as recognition word information to the utterance important word extraction apparatus.

  According to the important word extraction device in an utterance of the present invention, since the keyword list for specifying the important word is automatically created from the text information input by the user, it is possible to save labor for creating the keyword list. Play. In addition, since the text information input by the user often expresses the object of interest of the user itself, and it is considered that the erroneous input is less likely, the method of the present invention for obtaining a keyword from the text information uses the important word. It can be extracted with high accuracy.

  Further, according to the important word extraction system in an utterance of the present invention, since the relatively heavy processing speech recognition processing is shared by the recognized word extraction server via the network, the configuration of the important word extraction device in the utterance can be simplified. can do. As a result, the key word extraction device in the utterance can be made inexpensive and downsized.

The figure which shows the function structural example of the important word extraction apparatus 100 in an utterance of this invention. It is a figure which shows the operation | movement flow of the important word extraction apparatus 100 in an utterance, (a) shows the process after text input, (b) shows the process after speech input. The figure which shows the function structural example of the text input part 110. FIG. The figure which shows an example of the keyword list | wrist 140. FIG. The figure which shows the system configuration | structure of the important word extraction system 200 in an utterance of this invention. The figure which shows the operation | movement sequence of the important word extraction system 200 in an utterance. The figure which shows the system configuration | structure of the important word extraction system 300 in an utterance of this invention. The figure which shows an example of the external appearance of the important word extraction apparatus 320 in an utterance, and its utilization scene.

  Embodiments of the present invention will be described below with reference to the drawings. The same reference numerals are given to the same components in a plurality of drawings, and the description will not be repeated.

  FIG. 1 shows an example of the functional configuration of an apparatus for extracting important words in an utterance 100 of the present invention. The operation flow is shown in FIG. The utterance important word extraction device 100 includes a keyword list 140, a text input unit 110, a morpheme analysis unit 120, a word registration unit 130, a voice input unit 150, a voice recognition unit 160, and a key word extraction unit 170. And a control unit 180. The in-speech important word extraction device 100 is realized by reading a predetermined program into a computer composed of, for example, a ROM, a RAM, a CPU, and the like, and executing the program by the CPU. The same applies to other embodiments described below.

  Keywords are registered in the keyword list 140. Text information is input to the text input unit 110. FIG. 3 shows a functional configuration example of the text input unit 110. The text input unit 110 includes a keyboard 111, a tablet 112, a keyboard interface 113, a tablet interface 114, a control unit 180, and a cache memory 116. The keyboard 111, the tablet 112, the keyboard interface 113, and the tablet interface 114 are general ones. The control unit 180 is also a general unit configured by a CPU, a RAM, and a ROM when the utterance important word extracting apparatus 100 is configured by a computer, for example. The control unit 180 controls time-series operations and the like of each unit of the utterance important word extraction device 100. As a part of the control unit 180, a web browser (not shown) is started based on information input from the keyboard 111. The functions of the text input unit 110 that outputs text information by controlling operations such as browsing news and searching for information are shared. The cache memory 116 is a temporary file of the web browser.

  The text information output by the text input unit 110 includes text temporarily recorded in the cache memory 116 and text posted by a user on a blog or microblog. A specific example of the text information is, for example, a web screen title sentence temporarily recorded in the cache memory 116. The title sentence can be easily extracted by taking out the text with the HTML title tag. For example, it is assumed that “<title> Kagawa,“ next is different ”</ title>” that does not occur in Iraq is included in an HTML source of an article searched by inputting “Japan representative” as a search keyword of browser software. <Title> </ title> is a title tag, and a sentence surrounded by the title tags is a title sentence.

  Other than the title sentence, the entire text of the search keyword or article may be handled as text information. Text information is memorize | stored in memory | storage parts, such as RAM, of the computer which comprises the important word extraction apparatus 100 in an utterance.

  When text information is input from the keyboard 111 or the tablet 112 of the text input unit 110, the morpheme analysis unit 120 includes word and part of speech information obtained by dividing the text information output from the text input unit 110 into morphemes. A morpheme string is output (step S120). The control unit 180 determines whether text information is input to the text input unit 110 (Yes in step S180).

  The word registration unit 130 receives the morpheme string output from the morpheme analysis unit 120 as an input, and registers a word satisfying a predetermined condition in the keyword list 140 as a keyword (step S132). The predetermined condition is, for example, part of speech information. The word registration unit 130 uses only words having part-of-speech information of nouns and verbs as proper keywords, for example, according to predetermined conditions.

  The word registration unit 130 does not register a word whose part of speech information output from the morphological analysis unit 120 does not match a predetermined condition as a keyword (No in step S131). The processes of the morphological analysis unit 120 and the word registration unit 130 are repeated every time text information is input to the text input unit 110 until the text information ends (No in step S181).

  FIG. 4 shows an example of keywords registered in the keyword list 140. In the keyword list 140, for example, words representing only keywords such as “Shibuya” and “Hatsudai” may be registered as minimum necessary information. In addition to those words, an identifier (ID) for identifying a keyword may be given as shown in the first column of FIG. In addition, “registration opportunity”, “registration date / time”, “last use date / time”, and the like that registered the keyword may be registered. Here, “registration opportunity” is information indicating the type of keyword. “Initial registration” is a keyword that the utterance important word extraction apparatus 100 has in advance and is not deleted. “Text input” is a keyword registered from text information entered by a user on a blog, microblog, text memo, etc. at a certain date and time, and may be deleted after a predetermined time has elapsed. “Web browsing” is a type of keyword extracted from text information acquired by a user browsing the Internet.

  The keyword may be reset after elapse of a predetermined time measured based on “registration date” or “last use date”. When resetting a keyword, the important word extraction device 100 in the utterance includes a keyword list reset unit 190 indicated by a broken line in FIG.

  The keyword list resetting unit 190 resets the keywords individually after, for example, measuring the passage of a predetermined time from the “last use date”. Here, the elapse of the predetermined time may be, for example, a plurality of periods such as 24 hours, 2 days, 3 days, or a week later. For example, according to the information of “registration opportunity”, the period may be set according to the type. You can change it.

  Alternatively, for example, the keyword list 140 may be reset by a reset signal input from the outside in conjunction with an operation input for inputting text information from a user. The predetermined time in this case is an interval operated by the user, and is an irregular time. Alternatively, the keyword list reset unit 190 may be configured to output a reset signal at a predetermined time period. In this case, all the keywords registered in the keyword list 140 are deleted at a predetermined time interval. By resetting the keyword list 140 at such a predetermined time, the keyword list 140 composed of the latest important words can be kept.

  When speech information is input to the speech input unit 150 (Yes in step S183), the speech recognition unit 160 generates and outputs a text sentence composed of a word string divided into morphemes by performing speech recognition processing on the speech information ( Step S160). Audio information input to the audio input unit 150 is converted into an audio signal converted into a discrete digital signal at a sampling frequency of 16 kHz, for example. The speech recognition unit 160 obtains an acoustic feature amount by, for example, mel frequency cepstrum coefficient (MFCC) analysis for each frame in which a predetermined number (eg, 320) of discrete speech signals is one frame, and the acoustic likelihood and Generate and output a text sentence consisting of word strings divided into morphemes with the highest language likelihood. Also, the reliability of each word may be output simultaneously with the text sentence, and the reliability is calculated based on the posterior probability of the word in the N best candidates, for example, Reference 1 (Frank Wessel, Ralf Schluter, Klaus Macherey and Hermann Ney, “Confidence Measures for Large Vocabulary Continuous Speech Recognition”, IEEE Transactions on Speech and Audio Processing, Vol. 9, No. 3, March 2001.) may be used.

  The important word extraction unit 170 receives the text sentence output from the speech recognition unit 160 and extracts a word that matches the keyword as an important word (step S172). In addition, by using the reliability output by the voice recognition unit 160, only a recognized word that has a reliability equal to or higher than a certain threshold and that matches the keyword may be extracted as an important word. The processes of the voice recognition unit 160 and the important word extraction unit 170 are repeated every time voice information is input to the voice input unit 150 until the voice information ends (No in step S182).

  As described above, the utterance important word extraction apparatus 100 reads words included in the text information input to the text input unit, or words included in the text information acquired by browsing the Internet based on the text information. A keyword that can be registered as a keyword in the keyword list 140 and matches one of the keywords registered in the keyword list included in the voice information input to the voice input unit 150 is extracted as an important word.

  According to the utterance important word extraction apparatus 100, since an important word is extracted from voice information, an effect of facilitating creation of a personal activity history such as a life log or a memorandum can be achieved. In addition, since the keyword list for specifying the keyword is automatically created from the text information input to the text input unit, labor for creating the keyword list can be saved. In addition, text information input by the user often represents the object of interest of the user in many cases, and it is considered that there are few input errors, so that important words can be extracted with high accuracy.

  In the above-described embodiment, the example in which the part of speech of the word registered in the keyword list 140 as a keyword is “noun” or “verb” has been described, but the present invention is not limited to this example. For example, a word string connecting two or more types of parts of speech may be registered as a keyword. For example, a word string in which “nouns” and “suffixes” are continuous or a connection of three or more words may be registered as keywords to extract important words. By increasing the number of combinations of parts of speech that make up a keyword, it is possible to more specifically identify important words.

[Key words extraction system in utterance]
FIG. 5 shows a system configuration of an intra-speech key word extraction system 200 including the intra-speech key word extraction device 220 of the present invention. The utterance important word extraction system 200 includes a voice input terminal 210, an utterance important word extraction device 220, a network 240, and a recognized word extraction server 250. The operation of the utterance important word extraction system 200 will be described with reference to FIG.

  The voice input terminal 210 outputs a voice signal picked up by a microphone (not shown) (step S210) to the utterance important word extraction device 220 (step S211). The microphone is preferably a small one that can be worn on the chest of the user, for example, in order to record voices uttered by the user at all times. The microphone and the voice input terminal may be integrated. The form may be a tie pin, for example.

  As described above, the voice signal is a signal converted into a discrete digital signal at a sampling frequency of 16 kHz, for example, and is output to the in-speech important word extraction device 220 by wire or wirelessly. In the case of wireless, for example, Bluetooth (registered trademark) or wireless LAN, which is a short-range wireless technology that covers a distance of up to about 10 m called a wireless PAN (Personal Area Network), can be used.

[Key words extraction device]
The utterance important word extraction device 220 registers words satisfying a predetermined condition obtained from input text information as keywords, and a person from a voice file that records a voice signal output from the voice input terminal 210 as a voice file. The voice signal of the voice section obtained by cutting out only the voice part uttered and the voice start time information are transmitted to the recognized word extraction server 250 via the network 240, and the recognized word information that matches the keyword received from the recognized word extraction server 250. Are extracted as important words and output. Here, for example, a method described in Reference Document 2 (Japanese Patent Laid-Open No. 2012-48119) can be used to extract a voice section.

  With reference to FIG. 5, the operation of the keyword-in-speech extraction device 220 will be described in more detail. The important word extraction device 220 in the utterance includes a text input unit 221, a morphological analysis unit 222, a word registration unit 223, a keyword list 224, a voice recording unit 225, a voice segment extraction unit 226, and a voice transmission unit 227. , A recognized word information receiving unit 228, a keyword extraction unit 229, and a keyword display unit 230.

  Text input unit 221 receives text information (step S221). The morpheme analysis unit 222 outputs a morpheme string composed of word and part of speech information obtained by dividing the text information input to the text input unit 221 into a morpheme by performing morpheme analysis (step S222). The word registration unit 223 receives the morpheme string output from the morpheme analysis unit 222 as an input, and registers words satisfying a predetermined condition in the keyword list as keywords (step S223).

  The voice recording unit 225 records the voice signal sent from the voice input terminal 210 as a voice file (step S225). At this time, the recording start time, which is the time when the recording of the audio signal is started, is also recorded. The voice signal and the recording start time are stored in a storage unit such as a RAM of a computer constituting the important word extraction device 220 in the utterance.

  The voice segment extraction unit 226 creates a voice file by cutting out only a voice part uttered by a person from the voice file recorded by the voice recording unit 225, and outputs the voice file to the voice transmission unit 227 (step S226). At this time, the utterance start time information of the extracted audio file is also output to the audio transmission unit 227. The voice start time information is obtained from the elapsed time from the voice signal recording start time and the voice signal recording start time of the cut voice. That is, the voice start time information is obtained by adding the elapsed time from the voice signal recording start time of the cut voice to the voice signal recording start time. The method of cutting out only the audio part is well known as described above.

  The voice sending unit 227 transmits the voice signal of the voice section extracted by the voice section extraction unit 226 and the utterance start time information to the recognized word extraction server 250 via the network 240 (step S227). The recognized word information receiving unit 228 receives the recognized word information sent from the recognized word extraction server 250 via the network 240 (step S228). This recognition word information is the same as the text sentence output by the speech recognition unit 160 (FIG. 1).

  The keyword extraction unit 229 uses the recognition word information received by the recognition word information reception unit 228 as an input, and extracts a word that matches the keyword registered in the keyword list 224 as a keyword (step S229). The important word display unit 230 displays the important words extracted by the important word extraction unit 229 (step S230). The important word is displayed by display means such as a liquid crystal panel (not shown) provided in the utterance important word extraction device 220.

  The text input unit 221, the morpheme analysis unit 222, the word registration unit 223, the keyword list 224, the voice recording unit 225, the speech segment extraction unit 226, and the keyword extraction unit 229 of the important word extraction device 220 in the utterance are However, it is the same as each functional unit having the same name explained in the important word extracting device 100 (FIG. 1).

[Recognized word extraction server]
As shown in FIG. 5, the recognized word extraction server 250 includes a voice receiving unit 251, a voice recognizing unit 252, and a recognized word information transmitting unit 253. The voice receiving unit 251 receives a voice signal and utterance start time information transmitted from the intra-utterance important word extraction device 220 via the network 240 (step S251).

  The voice recognition unit 252 generates and outputs a text sentence composed of word string information obtained by dividing the voice signal received by the voice reception unit 251 into a morpheme through voice recognition processing (step S252). The recognized word information transmission unit 253 outputs the text sentence output by the voice recognition unit 252 and the utterance start time information received by the voice reception unit 251 to the important word extraction device 220 in the utterance via the network 240 (step S253). ).

  According to the intra-speech key word extraction system 200 described above, the relatively heavy processing speech recognition process is shared by the recognized word extraction server 250, so the intra-speech key word extraction device 100 has the same effect as the intra-speech key word extraction device 100. The configuration of the word extraction device 220 can be simplified. Further, by increasing the CPU power of the computer that implements the function of the recognized word extraction server 250, it is possible to perform voice recognition processing that is faster and more accurate than the voice recognition performance performed by the important word extraction device 100 in the utterance. .

  FIG. 7 shows a system configuration of an important word extraction system 300 in an utterance according to the present invention. The important word extraction system 300 in the utterance is different from the important word extraction system 200 in the utterance only in the configuration of the important word extraction device 320 in the utterance.

  The utterance important word extraction device 320 differs from the utterance important word extraction device 220 (FIG. 5) in that it includes an action history display unit 321. The action history display unit 321 displays an action history that is a combination of the important word extracted by the important word extraction unit 229 and the time zone in which the important word is extracted. The action history is displayed by display means such as a liquid crystal panel (not shown) provided in the utterance important word extraction device 320.

  FIG. 8 shows an example of the appearance and usage scenes of the voice input terminal 210 and the important word extraction device 320 in the utterance. The voice input terminal 210 is a tie pin type in the above example. The important word and the time zone in which the important word is extracted are displayed for each time zone on the display means of the important word extraction device 320 in the utterance. Thus, by displaying the important words and the time zone, the user can check the daily action history.

  Further, when the user taps an important word or time zone on the display means with a fingertip, the voice of the time zone recorded by the voice recording unit 225 may be reproduced by a speaker (not shown). . When reproducing the voice corresponding to the action history, the important word extraction device 320 in the utterance further includes an action history selection input unit 322 and a recorded data selection reproduction unit 323. The action history selection input unit 322 outputs a signal corresponding to an important word or time zone to the recording data selection / reproduction unit 323 in response to an input that a fingertip of the user selecting action history taps on the display unit.

  The recorded data selection / playback unit 323 reads the voice file recorded by the voice recording unit 225 based on the signal corresponding to the important word or time zone output from the action history selection input unit 322 and plays it back on the speaker. By making the audio file corresponding to the important word or time zone audible as audio, the user can look back on the action history in detail.

  In the utterance important word extraction apparatus and the utterance important word extraction system using the apparatus of the present invention described above, the text information that the user inputs to the utterance important word extraction apparatus is the object of interest of the user itself. Then, a word related to the text information is registered as a keyword, and a word matching the keyword is extracted as an important word. According to this method, since the keyword list can be automatically updated, the maintenance cost can be eliminated. Moreover, since it is considered that the text information input by the user contains almost no errors, it is possible to extract important words with high accuracy.

  The important word extraction device in an utterance of the present invention having such excellent effects can create a personal activity history such as a life log and a memorandum, and can provide a user with high convenience.

  When the processing means in the above apparatus is realized by a computer, the processing contents of the functions that each apparatus should have are described by a program. Then, by executing this program on the computer, the processing means in each apparatus is realized on the computer.

  The program is distributed by selling, transferring, or lending a portable recording medium such as a DVD or CD-ROM in which the program is recorded. Further, the program may be distributed by storing the program in a recording device of a server computer and transferring the program from the server computer to another computer via a network.

  Each means may be configured by executing a predetermined program on a computer, or at least a part of these processing contents may be realized by hardware.

Claims (7)

A keyword list with registered keywords,
A text input section for inputting text information;
A morpheme analysis unit that outputs a morpheme string composed of word and part-of-speech information obtained by morphologically analyzing the text information and dividing it into morphemes;
A word registration unit for registering a word satisfying a predetermined condition as a keyword in the keyword list using the morpheme string as an input;
A voice input unit for inputting voice information;
A speech recognition unit that generates and outputs a text sentence composed of a word string obtained by performing speech recognition processing on the speech information and dividing the speech information;
An important word extraction unit that extracts the word that matches any of the keywords registered in the keyword list as an important word, using the text sentence as an input;
An apparatus for extracting important words in an utterance. A voice input terminal,
An apparatus for extracting important words in an utterance connected to the voice input terminal;
Network,
A recognized word extraction server that communicates with the utterance key word extraction device via the network;
An important word extraction system in an utterance comprising
The voice input terminal outputs a voice signal picked up by a microphone to the important word extraction device in the utterance,
The utterance important word extraction device registers words satisfying a predetermined condition obtained from input text information as keywords, and only a voice portion uttered by a person from a voice file in which the voice signal is recorded as a voice file. The speech signal of the extracted speech section and the utterance start time information are transmitted to the recognized word extraction server via the network, and the word of the recognized word information that matches the keyword received from the recognized word extraction server Is extracted and output as
The recognized word extraction server includes a word string obtained by receiving a speech signal and speech start time information of a speech section from the important word extraction device in the utterance, and performing speech recognition processing on the speech signal of the speech section and dividing it into morphemes. A text sentence is generated, and the words constituting the text sentence are output as recognition word information to the important word extraction device in the utterance,
An important word extraction system in the utterance. An apparatus for extracting important words in an utterance used in the important word extracting system in an utterance according to claim 2,
A keyword list with registered keywords,
An audio recording unit that records audio signals sent from the audio input terminal;
A voice segment extraction unit that extracts a voice signal of a voice segment obtained by cutting out only a voice part uttered by a person from the voice file recorded in the voice recording unit, and adds generation start time information indicating the start time to the voice segment. When,
A voice sending unit that sends the voice signal of the voice segment extracted by the voice segment extraction unit and the utterance start time information to the recognition word extraction server via the network;
A text input section for inputting text information;
A morpheme analysis unit that outputs a morpheme string composed of word and part-of-speech information obtained by morphologically analyzing the text information and dividing it into morphemes;
A word registration unit for registering a word satisfying a predetermined condition as a keyword in the keyword list using the morpheme string as an input;
A recognition word information receiving unit for receiving recognition word information sent from the recognition word extraction server via the network;
A keyword extraction unit that extracts, as input, the recognition word information received by the recognition word information reception unit, as a keyword, a word that matches any of the keywords registered in the keyword list;
An apparatus for extracting important words in an utterance. In the utterance important word extraction device according to claim 1 or 3,
Furthermore,
An apparatus for extracting important words in an utterance, comprising: an action history display unit that displays an action history that includes the important word and a time zone in which the important word is extracted. A text input process in which text information is input;
A morpheme analysis process for outputting a morpheme sequence composed of word and part-of-speech information obtained by morphologically analyzing the text information and dividing it into morphemes;
Using the morpheme string as an input, a word registration process for registering a word satisfying a predetermined condition as a keyword in a keyword list;
A voice input process in which voice information is input;
A speech recognition unit that generates and outputs a text sentence composed of a word string obtained by performing speech recognition processing on the speech information and dividing the speech information;
An important word extraction process for extracting, as an important word, a word that matches any of the keywords registered in the keyword list, using the text sentence as an input,
A method for extracting important words in an utterance. A voice signal output process in which a voice input terminal outputs a voice signal picked up by a microphone to a key word extraction device in an utterance;
The speech important word extraction device records the voice signal as a voice file, and extracts the voice signal of the voice section obtained by cutting out only the voice part uttered by the person from the recorded voice file and the utterance start time information of the network. A voice transmission process to be transmitted to the recognized word extraction server via
The recognition word extraction server includes a word string obtained by receiving a speech signal and speech start time information of a speech section from the important word extraction device in the utterance, and performing speech recognition processing on the speech signal of the speech section and dividing it into morphemes. A recognition word information transmission process for generating a text sentence and outputting the words constituting the text sentence as recognition word information to the important word extraction device in the utterance;
The utterance important word extraction device registers a keyword obtained from text information input to a text input unit, and extracts a word of the recognized word information received from the recognized word extraction server that matches the keyword as an important word. The important word extraction process
A method for extracting important words in utterances.   A program for causing a computer to function as the important word extraction device in an utterance according to claim 1.