JP2004361769A - Voice authentication system and voice processing terminal - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a voice authentication system capable of easily recognizing whether a user is a licit user or not. <P>SOLUTION: The voice authentication system comprises adding the word data out of the item data 6 and word data 7 acquired from a mobile device 1 to a word dictionary 14 after establishing a wireless connection between the mobile device and a voice processing terminal 11 by a dictionary addition means 15; converting the voices inputted from the mobile device to text data by a voice data converting means 19; selecting the most approximate word from the words recorded in the word dictionary 14 by a voice recognizing means 17; and performing authentication by comparing the selected word and the word data added to the word dictionary 14 by an authentication word comparing means 18. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a voice recognition system and a voice processing terminal for recognizing whether a user is a valid user.
[0002]
[Prior art]
2. Description of the Related Art Conventionally, user authentication is performed to prevent an information terminal from being used by anyone other than the owner. As the user authentication method, an authentication method by inputting an owner ID or a password is generally used (for example, see Patent Document 1). Furthermore, in order to strengthen security, authentication methods using biometric information using fingerprints, voiceprints, irises, etc. are used, and measures are taken to prevent anyone other than the owner from using the information terminal. I have.
[0003]
[Patent Document 1]
JP-A-2002-133384
[Problems to be solved by the invention]
In the conventional authentication method by inputting an owner ID and a password, the owner ID and the password are easily stolen, and personal information cannot be protected in the case of theft, especially in a small and lightweight portable terminal. As for biometric information, biometric recognition using a portable information terminal with a small CPU power requires a complicated process and takes a long time to recognize. In addition, a scanner is provided in the portable information terminal to authenticate fingerprints and irises. There is a problem that must be done.
[0005]
An object of the present invention is to provide a voice authentication system and a voice processing terminal capable of easily recognizing whether a user is a valid user.
[0006]
[Means for Solving the Problems]
The voice authentication system according to claim 1 of the present invention is a voice authentication system including a mobile device and a voice processing terminal, wherein the mobile device stores item data and word data corresponding to the item data. Recording means, voice input means, and data input / output means for inputting / outputting the voice input from the voice input means and the item data and the word data. The mobile phone comprising: a data input / output unit wirelessly connected to the data input / output unit; a word dictionary in which words are recorded; a voice data conversion unit; a voice recognition unit; a dictionary addition unit; Adding the word data among the item data and the word data obtained from the mobile device after establishing a wireless connection with the device to the word dictionary by the dictionary adding means; The voice input from the portable device is converted into text data by the voice data conversion means, and the closest word is selected by the voice recognition means from the words recorded in the word dictionary. The authentication is performed by comparing the word data added to the word dictionary with the authentication word comparison means.
[0007]
According to a second aspect of the present invention, in the voice authentication system according to the first aspect, the voice processing terminal performs voice synthesis from the item data obtained from the mobile device after establishing a wireless connection with the mobile device. Speech synthesis means for generating data, outputting the generated speech synthesis data to the portable device, the portable device comprises speech data conversion means for converting the input speech synthesis data into speech, and speech output means, The converted voice is output from the voice output means.
[0008]
According to a third aspect of the present invention, there is provided a voice processing terminal, comprising: a data input / output unit wirelessly connected to a portable device; a word dictionary in which words are recorded; a voice data conversion unit; a voice recognition unit; And authentication word comparison means, and adds the first word data of the item data and the word data obtained from the portable device after establishing a wireless connection with the portable device to the word dictionary, and outputs the voice input from the portable device. Is converted to text data by the voice data conversion means, the closest word is selected by the voice recognition means from the words recorded in the word dictionary, and the selected word and the word added to the word dictionary are selected. The authentication is performed by comparing data with the authentication word comparing means.
[0009]
A voice processing terminal according to a fourth aspect of the present invention is the voice processing terminal according to the third aspect, wherein voice is generated from the item data obtained from the portable device after establishing a wireless connection with the portable device. And a synthesizing unit for outputting the generated speech synthesis data to the portable device.
[0010]
The voice authentication system according to claim 5 of the present invention is a voice authentication system including an IC card and a voice processing terminal, wherein the IC card stores item data and word data corresponding to the item data. Recording means, a common key, and data input / output means for inputting / outputting the common key, the item data, and the word data, wherein the voice processing terminal is wirelessly connected to the data input / output means of the IC card. Data input / output means, a common key, an encryption means for generating a random number and encrypting the generated random number with the common key, a data input / output means for wirelessly connecting to the data input / output means of the IC card, Random number checking means, word dictionary in which words are recorded, voice data converting means, voice recognizing means, dictionary adding means, voice output or display so that the user can recognize. A user recognition unit for displaying, a voice input unit, a voice recognition unit, and an authentication word comparison unit, wherein the voice processing terminal is encrypted by the encryption unit after establishing a wireless connection with the IC card. Outputting an encrypted random number to the IC card, the IC card decrypts the input encrypted random number with the common key provided in the IC card, and outputs the decrypted decrypted random number to the voice processing terminal; The voice processing terminal checks the input decrypted random number with the random number generated by the encryption unit and the random number check unit, and after the check, checks the word among the item data and the word data obtained from the IC card. Data is added to the word dictionary by the dictionary adding means, and the voice input from the IC card is converted into text data by the voice data converting means, The closest word is selected from the words recorded in the word dictionary by the voice recognition means, the selected word is made to be recognized by the user by the user recognition means, and the voice inputted from the voice input means is converted into the voice. The method is characterized in that the recognition is performed by the recognition unit, and the recognized word is compared with the word data added to the word dictionary by the authentication word comparison unit to perform authentication.
[0011]
The voice processing terminal according to claim 6 of the present invention is a data input / output unit that wirelessly connects to an IC card, a common key, an encryption unit that generates a random number and encrypts the generated random number with the common key, Data input / output means wirelessly connected to the data input / output means of the IC card, random number confirmation means, word dictionary in which words are recorded, speech data conversion means, speech recognition means, dictionary addition means, A user recognition means for outputting a voice or displaying on a display so that a user can recognize the information, a voice input means, a voice recognition means, and an authentication word comparison means, and the encryption means after establishing a wireless connection with the IC card. And outputs the decrypted random number input from the IC card to the random number generated by the encryption means of the IC card and the random number confirmation means. After the confirmation, the word data of the item data and the word data obtained from the IC card is added to the word dictionary by the dictionary adding means, and the voice input from the IC card is converted into text data by the voice data converting means. The closest word from the words recorded in the word dictionary after conversion is selected by the voice recognition means, the selected word is recognized by the user by the user recognition means, and input by the voice input means. The voice is recognized by the voice recognition unit, and the recognized word is compared with the word data added to the word dictionary by the authentication word comparison unit to perform authentication.
[0012]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, embodiments of the present invention will be described with reference to the drawings.
[0013]
FIG. 1 is a diagram showing functional blocks of the voice authentication system according to the first embodiment of the present invention. In this embodiment, a case where a headset with a wireless function is used as a portable device is described. 1 is a headset, 2 is a microphone, 3 is an earphone, and a user of the headset 1 inputs a sound from the microphone 2 and listens to a sound from the earphone 3. Reference numeral 4 denotes an audio data processing unit, and reference numeral 5 denotes a wireless communication unit. The audio data processing unit 4 converts the sound input from the microphone 2 into sound data, and further converts the sound into data that can be transmitted by the wireless communication unit 5.
[0014]
Reference numeral 9 denotes an authentication information recording unit, in which item data 6 and word data 7 corresponding to the item data 6 are recorded. Reference numeral 8 denotes an authentication information conversion unit that converts the item data 6 and the word data 7 into data to be transmitted wirelessly, and passes the data to the wireless transmission unit 5.
[0015]
The wireless communication unit 5 is a communication unit that performs communication within a short distance, for example, within several meters, and automatically starts a connection process when a communicable wireless communication unit is within a connectable range. .
[0016]
FIG. 2 is a diagram showing an example of item data 6 and word data 7 recorded in the authentication information recording means 9 of FIG. The item data 6 is a content to be asked by the voice processing terminal, which will be described later, and records items to which the owner of the headset 1 can answer, such as “favorite animal” and “favorite place”. In the word data 7, a word corresponding to the item data 6 is recorded. For example, if a favorite animal of the owner of the headset 1 is lost, “Dragon” is recorded. In addition, phonetic symbols for voice recognition and the like may be recorded. These authentication data can be written and read only by a trusted information terminal that has been authenticated with the headset and can be easily and safely changed by a legitimate user.
[0017]
In FIG. 1, reference numeral 11 denotes a voice processing terminal, and the voice processing terminal 11 is a terminal capable of mounting voice recognition and voice synthesis functions, such as a personal digital assistant, a personal computer, and a voice recognition / synthesis dedicated to CPU and memory. Anything such as a terminal is acceptable. A wireless communication unit 12 is connected to the wireless communication unit 5 of the headset 1 to transmit and receive data. Reference numeral 13 denotes a data conversion unit that extracts voice data, item data 6, and word data 7 from the received data input from the wireless communication unit 12. Further, the audio data is converted into data to be transmitted by the wireless communication unit 12. Reference numeral 14 denotes a dictionary for speech recognition, and reference numeral 15 denotes a dictionary addition / deletion unit that adds or deletes the word data extracted by the data conversion unit 13 to / from the dictionary 14 for speech recognition.
[0018]
FIG. 3 is a diagram showing an example of the word 14a registered in the speech recognition dictionary 14 of FIG. In the speech recognition dictionary 14, a plurality of words 14a are registered in advance. The words to be registered are not limited to Japanese and English. Also, reading symbols and the like may be recorded in addition to words. (A) shows an example of the word data 7 and the words 14a registered in the speech recognition dictionary 14 before the word data 7 is added. (B) shows a word 14b obtained by adding the word data 7 to the speech recognition word dictionary 14. The position where the word data is added is not particularly limited, and may be anywhere. The speech recognition dictionary 14 to which the word data 7 has been added is used for speech recognition.
[0019]
In FIG. 1, reference numeral 16 denotes a voice synthesizing unit, which converts the item data 6 extracted by the data converting unit 13 into voice data. A voice recognition unit 17 converts the voice data extracted by the data conversion unit 13 into characters, and selects the closest word from the words recorded in the voice recognition dictionary 14. Note that the voice recognition unit 17 may not select a word from the voice recognition dictionary 14 when there is no word close to the recognized word in the voice recognition dictionary 14. Reference numeral 18 denotes a recognized word comparing unit that compares the word selected by the voice recognizing unit 17 with the word data from the data converting unit 13 to perform recognition.
[0020]
FIG. 4 is a diagram showing an operation flow of the configuration of FIG. The same parts as those in FIG. 1 are given the same numbers.
[0021]
When the headset 1 and the audio processing terminal 11 approach a distance at which the wireless communication means 5 and 12 can communicate with each other, the headset 1 and the audio processing terminal 11 perform wireless connection processing and establish a wireless connection 101 for wireless communication. Next, the voice processing terminal 11 transmits a request 102 for authentication information to the headset 1. The authentication information here indicates the item data 6 and the word data 7 in FIG. The headset 1 that has received 103 the authentication information request transmits 104 the recorded authentication information.
[0022]
The voice processing terminal 11 that has received the authentication information 105 adds the received word data 7 to the voice recognition dictionary 14 by the dictionary addition / deletion unit 15. Further, the voice synthesis means 16 performs voice synthesis processing 107 for generating voice synthesized data from the item data 6 and transmits the generated voice synthesized data 108. Here, the generated speech synthesis data may be only the item data, or may be speech synthesis data in which a question is added to the item data, for example, a word such as “what?”.
[0023]
The headset 1 receiving the voice synthesis data 109 converts the voice into voice 110 by the voice data processing means 4, and outputs voice from the earphone 3. The user wearing the headset 1 utters the answer to the question by the sound heard through the earphone 3. The voice uttered by the user is input from the microphone 2 to the headset 1, converted 111 by the voice data processing unit 4 into voice data for recognition for wireless transmission, and transmitted 112 as voice data for recognition.
[0024]
The speech processing terminal 11 that has received the recognition speech data 113 converts the speech data into text data by the speech data conversion means 19, and the speech data input from the words recorded in the speech recognition word dictionary 14 by the speech recognition means 17. Select the word closest to. Next, the recognized word comparison unit 18 compares the word data 7 added to the speech recognition word dictionary 14 with the word selected by the speech recognition unit 17 115. The user is regarded as a valid user, and makes the voice processing terminal 11 usable. If the previously received word data 7 does not match the word selected by the voice recognition means 17, use of the voice processing terminal as an unauthorized user is disabled.
[0025]
The operation flow of performing voice recognition once has been described above. However, when a plurality of item data 6 and voice data are registered, the operation flow of FIG. 4 is repeated a plurality of times, and stricter authentication may be performed. Even if the voice is wrong by the owner, even if the question is asked several times, the voice processing terminal may be used if the voice is correct even once.
[0026]
According to the voice authentication system of the present embodiment described above, even if there is no information for authentication on the voice processing terminal side first, the information from the headset side can easily use the proper owner by voice using the voice. Can be determined. In addition, by providing a dictionary in advance on the voice processing terminal side, a plurality of selectable words can be set, and it is possible to cope with accidental authentication using noise or close words.
[0027]
In addition, in cases other than the initial state, writing of item data and word data in the headset is enabled only when the voice authentication of the present embodiment is confirmed, so that unauthorized registration by an unauthorized user cannot be performed. Become like
[0028]
Also, in the present embodiment, when the headset and the audio processing terminal are wirelessly connected, the device can be authenticated by authenticating each other with a unique ID such as a terminal ID.
[0029]
FIG. 5 is a diagram showing functional blocks of the voice authentication system according to the second embodiment of the present invention. The same parts as those in FIG. 1 are given the same numbers.
[0030]
The voice authentication system in FIG. 5 is an example in which an IC card 21 is used for recording and communicating item data 6 and word data 7 instead of the headset 1 in FIG. Item data 6 and word data 7 are recorded on the IC card 21, and the authentication information conversion means 25 converts the data into data conforming to the communication standard of the IC card 21. The wireless communication unit 24 transmits the input / output data of the authentication information conversion unit 25 to the voice processing terminal 31. Further, the IC card 21 has a common key 22 and a decryption unit 23 for decrypting the encrypted data using the common key.
[0031]
The voice processing terminal 31 has a microphone 37 as voice input means and a speaker 36 as voice output means. The voice from the microphone 37 is connected to the input of the voice recognition means 17 and the speaker 36 is connected to the output of the voice synthesis means 16. In addition, the voice processing terminal 31 includes a common key 32 for authenticating the IC card 21, an encryption unit 34 for encrypting the random number generated by the random number generation unit 33, and a random number generated by the random number generation unit 33 and the IC card 21. There is a random number data confirmation unit 38 for comparing the returned random numbers.
[0032]
FIG. 6 is a diagram showing an operation flow of the configuration of FIG. The same parts as those in FIG. 4 are given the same numbers.
[0033]
The IC card 21 and the voice processing terminal 31 establish a wireless connection 101 as in FIG. After that, the authentication of the IC card 21 is performed. Therefore, the voice processing terminal 31 generates 201 a random number by the random number generation unit 33, performs a random number encryption 202 using the common key 230, and transmits 203 the encrypted random number to the IC card 21. The IC card 21 that has received the encrypted random number 204 decrypts the encrypted random number with the common key 231 and transmits 206 the decrypted random number.
[0034]
The voice processing terminal 31 that has received 207 the decrypted random number compares the decrypted random number with the previously generated random number 208, and if they match, authenticates that the IC card 21 is valid and continues the operation. If it is determined that the random numbers do not match and the IC card is an unauthorized IC card, the wireless connection is released.
[0035]
Hereinafter, the operations from steps 102 to 106 described in FIG. 4 are performed, and then a voice synthesis process 107 for generating voice synthesized data from the item data 6 is performed. Pour as The answer to the question uttered by the user who heard the question by the voice heard from the speaker 36 is subjected to voice recognition 114 by the voice recognition means 17 and then added to the voice recognition word dictionary 14 first by the recognized word comparison means 18. The word data 7 thus obtained is compared 115 with the word selected by the voice recognition means 17, and if they match, the user of the headset 1 is regarded as a valid user, and the voice processing terminal 11 can be used. If the previously received word data 7 does not match the word selected by the voice recognition means 17, use of the voice processing terminal as an unauthorized user is disabled.
[0036]
In this embodiment, since the IC card 21 is used, the IC card 21 can be used for opening and closing a door as an IC card key capable of authenticating the user, in addition to a usage method such as enabling terminal operation of the voice processing terminal 31. it can.
[0037]
FIG. 7 is a diagram illustrating functional blocks of the voice authentication system according to the third embodiment. The same components as those in FIG.
[0038]
FIG. 7 shows a case where the voice processing terminal 61 does not have voice synthesis and voice output means. Instead, the voice processing terminal 61 is provided with a display data generation means 62 and a display 63. Is displayed on the display 63.
[0039]
FIG. 8 is a diagram showing functional blocks of the voice authentication system according to the third embodiment of the present invention, and is a diagram showing an example of item data 64 displayed on the display 63 of FIG. The user of the IC card 21 looks at the display 63 and answers with a voice. Hereinafter, the operation flow is the same as that of FIG.
[0040]
According to this embodiment, the voice synthesis function of the voice processing terminal can be omitted, and the voice processing terminal can be realized with less memory.
[0041]
In the above embodiment, a wireless type, that is, a non-contact type IC card has been described as an example, but an input / output terminal is used instead of the wireless communication unit 24 in FIGS. 5 and 7, and an IC card is used instead of the wireless communication unit 35. By using the card reader / writer means, a contact type IC card can be used.
[0042]
【The invention's effect】
According to the present invention, it is possible to obtain a voice authentication system and a voice processing terminal capable of easily recognizing whether a user is a valid user.
[Brief description of the drawings]
FIG. 1 is a diagram showing functional blocks of a voice authentication system according to a first embodiment of the present invention.
FIG. 2 is a diagram showing an example of item data and word data recorded in an authentication information recording unit of FIG. 1;
FIG. 3 is a diagram showing an example of words registered in the voice recognition dictionary of FIG.
FIG. 4 is a diagram showing an operation flow of the configuration of FIG. 1;
FIG. 5 is a diagram showing functional blocks of a voice authentication system according to a second embodiment of the present invention.
FIG. 6 is a diagram showing an operation flow of the configuration of FIG. 5;
FIG. 7 is a diagram showing functional blocks of a voice authentication system according to a third embodiment of the present invention.
8 is a diagram showing an example of item data displayed on the display of FIG. 7;
[Explanation of symbols]
1: headset, 2: microphone, 3: earphone, 4: audio data processing means, 5: wireless communication means, 6: item data, 7: word data, 8: authentication information conversion means, 9: authentication information recording means, 11: voice processing terminal, 12: wireless communication means, 13: data conversion means, 14: dictionary for voice recognition, 15: dictionary addition / deletion means, 16: voice synthesis means, 17: voice recognition means, 18: authentication word Comparison means, 21: IC card, 22: common key, 23: decryption means, 24: wireless communication means, 25: authentication information conversion means, 31: voice processing terminal, 32: common key, 33: random number generation means, 34 : Encryption means, 35: wireless communication means, 36: speaker, 37: microphone, 61: audio processing terminal, 62: display data creation means, 63: display.

Claims (6)

A voice authentication system including a mobile device and a voice processing terminal,
The portable device includes a recording unit in which item data and word data corresponding to the item data are recorded, a voice input unit, and inputs and outputs a voice input from the voice input unit, the item data, and the word data. Data input / output means
The voice processing terminal is a data input / output unit that wirelessly connects to the data input / output unit of the portable device, a word dictionary in which words are recorded, a voice data conversion unit, a voice recognition unit, a dictionary addition unit, An authentication word comparison unit, wherein the word data among the item data and the word data obtained from the mobile device after establishing a wireless connection with the mobile device is added to the word dictionary by the dictionary addition unit, The voice input from the portable device is converted into text data by the voice data conversion means, and the closest word is selected by the voice recognition means from the words recorded in the word dictionary, and the selected word and the word A voice authentication system, wherein the authentication is performed by comparing the word data added to the dictionary with the authentication word comparison means. 2. The voice authentication system according to claim 1, wherein the voice processing terminal includes voice synthesis means for generating voice synthesis data from the item data obtained from the mobile device after establishing a wireless connection with the mobile device, and the generated voice. Outputting the synthesized data to the portable device, the portable device including a voice data converting unit and a voice output unit for converting the input voice synthesized data into voice, and outputting the converted voice from the voice output unit. A speech recognition system characterized by the following. A data input / output unit that wirelessly connects to the mobile device, a word dictionary in which words are recorded, a voice data conversion unit, a voice recognition unit, a dictionary addition unit, and an authentication word comparison unit; After establishing the wireless connection, the word data among the item data and word data obtained from the portable device is added to the word dictionary, and the voice input from the portable device is converted into text data by the voice data converting means. The closest word is selected from the words recorded in the word dictionary by the voice recognition means, and the selected word is compared with the word data added to the word dictionary by the authentication word comparison means for authentication. A voice processing terminal characterized by performing: 4. The voice processing terminal according to claim 3, further comprising: voice synthesis means for generating voice synthesized data from the item data obtained from the mobile device after establishing a wireless connection with the mobile device, wherein the generated voice synthesized data is transmitted to the mobile device. An audio processing terminal for outputting to a device. A voice authentication system including an IC card and a voice processing terminal,
The IC card includes: a recording unit in which item data and word data corresponding to the item data are recorded; a common key; and a data input / output unit that inputs and outputs the common key, the item data, and the word data. Prepare,
The voice processing terminal includes: a data input / output unit wirelessly connected to a data input / output unit of the IC card; a common key; an encryption unit configured to generate a random number and encrypt the generated random number with the common key; A data input / output unit wirelessly connected to the data input / output unit of the IC card, a random number check unit, a word dictionary in which words are recorded, a voice data conversion unit, a voice recognition unit, a dictionary addition unit, a user User recognition means for voice output or display display for recognition, voice input means, voice recognition means, authentication word comparison means,
The voice processing terminal outputs an encrypted random number encrypted by the encryption unit to the IC card after establishing a wireless connection with the IC card,
The IC card decrypts the input encrypted random number with the common key provided in the IC card, and outputs the decrypted decrypted random number to the voice processing terminal.
The voice processing terminal checks the input decrypted random number with the random number generated by the encryption unit and the random number check unit, and after the check, checks the word among the item data and the word data obtained from the IC card. Data is added to the word dictionary by the dictionary adding unit, and the voice input from the IC card is converted into text data by the voice data converting unit, and the closest word from the words recorded in the word dictionary is converted. The word selected by the voice recognition unit, the selected word is recognized by the user by the user recognition unit, and the voice input from the voice input unit is voice-recognized by the voice recognition unit. A voice authentication system, wherein the authentication is performed by comparing the word data added to the dictionary with the authentication word comparison means. Data input / output means for wirelessly connecting to the IC card, a common key, encryption means for generating a random number and encrypting the generated random number with the common key, and data wirelessly connected to the data input / output means of the IC card Input / output means, random number confirming means, word dictionary in which words are recorded, voice data converting means, voice recognizing means, dictionary adding means, and use of voice output or display for user recognition. User recognition means, voice input means, voice recognition means, and authentication word comparison means, and after establishing a wireless connection with the IC card, outputs an encrypted random number encrypted by the encryption means to the IC card. Then, the decrypted random number input from the IC card is confirmed by the random number confirming means and the random number generated by the encrypting means of the IC card, and after confirming, it is obtained from the IC card. The word data among the item data and the word data is added to the word dictionary by the dictionary adding unit, and the voice input from the IC card is converted into text data by the voice data converting unit and recorded in the word dictionary. The closest word is selected by the voice recognition means, the user recognizes the selected word by the user recognition means, and the voice input by the voice input means is recognized by the voice recognition means. A voice recognition terminal for performing authentication by comparing the word recognized by voice with the word data added to the word dictionary by the authentication word comparison means.

Images