JP2012244400A - Telephone apparatus and control method thereof - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a telephone apparatus which has automatic answering and speech analyzing functions, and its control method.SOLUTION: The telephone apparatus includes: line connection control means of controlling a call connection with a telephone line; storage means of saving at least speech information as digital data; speech reproduction means of converting the speech data saved in the storage means into analog data and reproducing and outputting the data; speech recording means of converting a speech of a telephone call into digital data and saving the data in the storage means; speech analysis means of extracting information on one or more of the loudness, frequency, speed, pitch, and formant of a sound from the speech of the telephone call, and comparing it with predetermined stored patterns to determine how much the caller is stressed; input/output means of making the respective means function through user's operation and outputting a result thereof; and control means of managing cooperation among and control over the respective means.

Description

  The present invention relates to a telephone device and a control method thereof, and more particularly to a telephone device having an automatic response and a voice analysis function and a control method thereof.

  There is no decrease in fraudulent acts by telephone, and so-called call-type wire fraud (named "Oleore fraud") targeting elderly people with particularly low judgment is reported as a new fraudulent act with various methods changed. As always.

  Measures against scams include refusing phone numbers other than pre-registered phone numbers, refusing calls from non-notified / public phones, displaying alert messages or receiving voice messages when receiving calls, and voice recognition during calls. It is known to compare with a conversation pattern or keyword dictionary registered in advance by converting to text, and to collate with a pre-registered voiceprint database. Patent Document 1 discloses a crime prevention guidance notification system. 2 discloses a topic determination device and a topic determination method.

JP 2010-273071 A JP 2010-197706 A

  However, in the configuration for collating with the database of keywords and voiceprints as described above, not only a large-scale storage capacity is required for the device or system, but also a database by an external server based on a wide range and a large amount of crime history is provided. There is a problem of necessity.

  The present invention has been made in view of the above-described conventional problems, and an object of the present invention is to provide an automatic response and voice that can suppress telephone fraud with a relatively simple configuration without requiring a large database. An object of the present invention is to provide a telephone device having an analysis function and a control method thereof.

  In order to achieve the above object, a telephone device of the present invention includes a line connection control means for performing call connection control with a telephone line, a storage means for storing at least voice information as digital data, and stored in the storage means. Voice reproduction means for analogly converting the reproduced voice data to be reproduced and output, voice recording means for converting voice data during a call to digital data and storing the data in the storage means, and the volume, frequency, One or more pieces of information of speed, pitch, and formant are extracted and compared with a predetermined storage pattern, and voice analysis means for judging the degree of stress of the caller, and each means based on user operation And input / output means for outputting the result, and control means for controlling cooperation and control of the means.

In the telephone device of the present invention, the voice analysis means performs a Fourier transform on the voice of the other party during the call to extract a frequency power spectrum, and the sound volume, frequency, speed, pitch, formant from the extracted voice component The degree of stress is determined based on a change of any one or more of information in a predetermined time, and it is determined that the stress is detected when the change of the corresponding component exceeds a predetermined change amount and the number of changes. You may make it do.
Preferably, when the voice analysis means detects stress, the control means calls a telephone number registered in advance by the line connection control means to perform a three-party call.
When the voice analysis means detects stress, the telephone device of the present invention further comprises a mail transmission means for notifying that the stress has been detected to a pre-registered mail address via a communication network. Is preferred.
The telephone device of the present invention preferably further comprises external input / output means for recording voice data stored in the storage means on a memory card connected externally.

  The telephone device control method of the present invention made to achieve the above object is a control method in a telephone device having an automatic answering and voice analysis function. Call if a phone call is detected, and a response is sent to the other party when a non-registered phone number is sent. Sounding a sound to start the call and storing the voice information during the call as digital data, and information of one or more of loudness, frequency, speed, pitch, formant from the voice during the call Extracting the data and comparing it with a predetermined memory pattern to determine the stress level of the other party and detecting the stress, and when detecting the stress, a pre-registered mail address The detection of the said stress addressed and having a the steps of causing the three-way call with a call to pre-registered telephone number and notifies by e-mail.

  According to the present invention, an unregistered phone number is rejected, a voice message to the effect that the other party is recorded is automatically transmitted, and when a stress is detected from the other party's voice during a call, an email is transmitted to the registered party. Since a third party can participate in a conversation through a three-party call, telephone fraud can be suppressed with a relatively simple configuration without requiring a large reference database as in the prior art. Further, the telephone device of the present invention can be realized by a configuration in which it is inserted as a telephone connection device (system BOX) between a commercially available telephone and a telephone line, so that there is an advantage that it is easy to introduce. Furthermore, since it can be incorporated into a mobile phone or a CTI server as an application program, it can be realized in various forms.

It is a schematic block diagram of the telephone apparatus by one Embodiment of this invention. It is a processing flowchart which shows the control method of the telephone apparatus by one Embodiment of this invention. It is a schematic block diagram of the telephone connection apparatus by one Embodiment of this invention. It is explanatory drawing of the telephone connection apparatus shown in FIG.

  Hereinafter, a specific example of a mode for carrying out the telephone device of the present invention will be described in detail with reference to the drawings.

  FIG. 1 is a schematic configuration diagram of a telephone device according to an embodiment of the present invention.

  Referring to FIG. 1, the telephone device 1 of the present embodiment includes a control unit 11, a storage unit 12, a voice recording unit 13, a voice reproduction unit 14, a voice analysis unit 15, an input / output unit 16, an external input / output unit 17, and a line. Connection control means 21 and mail transmission means 31 are provided.

  The control means 11 includes a main CPU (central processing unit), and plays a role of overall processing and control so as to function as the telephone device 1 in cooperation with each means and other devices.

  The storage means 12 is composed of a ROM (Read Only Memory), a RAM (Random Access Memory), a Flash memory, or an SSD (Solid State Drive), etc., storing programs for operating each means, and voice data during a call. Recording, storing a voice message to be sent to the called party, and the like are read by the control means and used by each means.

  The voice recording means 13 converts the voice during a call into digital data by an A / D converter (not shown), compresses the data into ADPCM, MP3, WMA, AAC, ATRAC, etc., and saves it in the storage means 12.

  The audio reproduction means 14 decodes the compressed audio digital data stored in the storage means 12, converts it into analog information by a D / A converter (not shown), and reproduces and outputs the audio. The voice data recorded in the storage means 12 during the call is reproduced, and the automatic response message to the call destination registered in the storage means 12 and the voice message to the user are reproduced.

  The voice analysis means 15 is used for various changes in the voice of the speaker peculiar to the fraud, such as a feeling of ups and downs, a change in voice color or speaker, or suddenly screaming loudly. Each feature pattern is extracted in advance from the voice information of the other party during the call using the specific information of the sound, such as the loudness, frequency, speed, pitch, formant, etc. The stress level of the caller is determined by comparing with a predetermined storage pattern registered in the storage unit 12. The degree of stress is determined based on the change in the predetermined time for the extracted information of each voice.

  Various speech analysis methods are known. In this embodiment, the speech speed is determined based on the sound volume and speech recognition, and the frequency power spectrum is obtained by Fourier transform (FFT) to determine the frequency of the speech. Extract components, pitch components, and formants. In these processes, normalization processing for aligning the entire volume of the sound of the sound data to a predetermined amplitude, or filtering the sound data subjected to Fourier transform and then performing inverse Fourier transform (IFFT) again to extract a sound waveform on the time axis. Cepstrum analysis is also included. For each piece of extracted audio information, it is detected that the change in the corresponding component at a predetermined time has exceeded a predetermined change amount and change count set in advance, and the detection count set for each piece of information is used as a basis. It is determined that stress is detected comprehensively. Such information may be further obtained for a predetermined frequency band component. The details of the extraction method of each component such as sound volume, frequency, speed, pitch, formant, etc. can be used here, and the description thereof is omitted here.

  The input / output means 16 is an off-hook, on-hook, 0-9, *, #, etc. dial / push button when using the telephone device 1, a microphone / speaker / transmitter / receiver, a display LED, LCD (liquid crystal display), etc. Are used for dialing, telephone number registration, voice message recording / playback, various input / output operations, message display according to operation, warning display, and the like.

  The external input / output unit 17 is used for recording the audio data stored in the storage unit 12 in an externally connected memory card, and has a memory card slot and a USB interface. Note that details and description of other hardware constituting the telephone device 1 are omitted.

  The line connection control means 21 is responsible for call connection control of the telephone device 1 connected to the telephone line 2 such as PSTN (Public Switched Telephone Network), and allows the user and the other party to talk through the handset. When an incoming call is received, the control means 11 is controlled so that an automatic response message for confirming whether or not the call is registered in the storage means 12 is sent to the other party via the voice reproduction means 14 before the call is started.

  Further, the line connection control means 21 is controlled by the control means 11 so that a three-party call is possible when the voice analysis means 15 detects a stress. The three-party call is registered in advance with the telecommunications company that owns the telephone line 2 and is set by the telecommunications company by notifying by dialing when using it, or the telephone device with two independent communication lines prepared 1 can be realized by providing a pseudo exchange function.

  When the voice analyzing means 15 detects the stress, the mail transmitting means 31 indicates that the stress is detected to the mail address registered in advance in the storage means 12 via the communication network 3 prepared separately from the telephone line 2. Notify by email. The communication network 3 can be shared with a telephone line by using an ADSL modem.

  Each means such as the voice recording means 13, the voice reproduction means 14, and the voice analysis means 15 is controlled by the control means 11 by executing a program recorded in the ROM, the flash memory, or the SSD of the storage means 12, and the A / D It can be made to function by cooperating with a converter, a D / A converter, or a device such as a DSP or ASIC (not shown).

  FIG. 2 is a process flow diagram illustrating a method for controlling a telephone device according to an embodiment of the present invention.

  Referring to FIG. 2, in the control method for telephone device 1 according to the present embodiment, when control means 11 first detects an incoming call via line connection control means 21 (step S101), the other party's telephone number remains silent. Is a telephone number registered in the storage means 12 (step S102). If the telephone number is registered in advance and permitted, the telephone rings and the normal operation is performed by the user's off-hook operation. The process shifts to conversation (step S104), and after the conversation is finished, the handset is placed by the user's on-hook operation and the process is normally terminated (step S105).

  In step S103, when the destination telephone number is a non-notification, a public telephone, or a telephone number that is not registered in the storage means 12, the control means 11 is in a call that has been registered in the storage means 12 in advance. A voice message asking whether or not the voice can be recorded is sent to the other party via the voice reproducing means 14 by the automatic response function, and a reply by dial tone from the other party is waited (step S016).

  When a dial tone for recording disapproval is received or there is no response before the predetermined time has elapsed, the control means 11 sends a rejection message registered in the storage means 12 to the other party via the voice reproduction means 14 ( In step S108), the communication line is disconnected through the line connection control means 21, and the communication line is forcibly terminated without making a call (step S110). At this time, it is registered in the rejection list of the storage means 12 (step S109). When the telephone number is notified, it is recorded together with the telephone number, in the case of non-notification or a public telephone, and the date and time of incoming call are recorded. You may make it possible to take out to a memory card etc. via this.

  When a dial tone for accepting recording is received in step S107, the control means 11 sends a voice message indicating that recording is registered in the storage means 12 to the other party via the voice reproduction means 14 (step S107). S111), a ringing tone of the telephone is sounded and a call with the other party is started by the user's off-hook operation. The call voice of the other party is analyzed to determine the degree of stress (step S112). For each piece of information extracted from the voice of the other party during the call, the control means 11 performs stress based on the number of detections in which the change of the corresponding component in a predetermined time exceeds a predetermined change amount and the number of changes. Is detected (step S113), and if no stress is detected during a call, the handset is placed by the user's on-hook operation after the conversation is over, and the process ends normally (step S114). You may abbreviate | omit the rejection message of step S108 sent to an other party, and the voice message to the effect of starting recording of step S111. Alternatively, instead of the voice message to the other party in step S111, a voice message alerting the user may be announced before starting the conversation, as will be described later.

  In step S113, when stress is detected in the voice of the other party during the call, the control unit 11 detects the stress via the communication network 3 to the mail address registered in advance in the storage unit 12 by the mail transmission unit 31. This is notified by e-mail, and after the conversation is finished, the handset is placed by the user's on-hook operation, and the process ends (step S117). Further, when stress is detected in the voice of the other party during a call, a dial tone for performing a three-party call is transmitted to the communication company that owns the telephone line 2 via the line connection control means 21 and the storage means 12 is stored in advance. It is also possible to call a telephone number registered in the website to encourage participation in a third party conversation. (Step S115).

  At the end of the conversation, the control means 11 registers the incoming date and time together with the destination telephone number in the rejection list of the storage means 12 (step S116) so that it can be taken out to a memory card or the like via the external input / output means 17. Also good.

  FIG. 3 is a schematic configuration diagram of a telephone connection device according to an embodiment of the present invention, and FIG. 4 is an explanatory diagram of the telephone connection device shown in FIG.

  3 and 4 show a telephone apparatus in which a telephone connection device 4 is provided as a system BOX between a commercially available telephone 5 and a communication line 2, and some of the telephone 5 is attached to the outside. Although there is a difference in function, the overall configuration as a telephone device is substantially the same as that of the telephone device 1 described above, and therefore, a duplicate description in the telephone connection device 4 is omitted.

  Referring to FIGS. 3 and 4, in the telephone connection device 4 of the present embodiment, a communication line such as PSTN and a telephone 5 that is generally marketed are connected to the external interface 42. A plurality of external interfaces 42 to which the telephone 5 is connected may be prepared. When this apparatus is introduced, first, the telephone numbers of family members, friends and acquaintances are registered in the storage means 12.

  When there is an incoming call via the communication line 2, the telephone connection device 4 does not send a ringing signal to the telephone 5 and checks the telephone number registered in the storage unit 12 without ringing. Via means 14, “Your phone number is not registered. If you wish to make a call, press the number 1 on the push button after the beep. However, the conversation will be recorded and saved at the same time. The data will be generated and registered in this system. If it is still OK, please press No. 1. Then, please send an announcement and "Peep" to the other party. The setting of whether or not to receive a ringing or no ringing when there is an incoming call may be changed by a dip switch (not shown) provided in the telephone connection device 4 or a push button operation of the telephone 5. . In addition, even in the case of unregistered phone numbers within the same area, such as the same area code or the same station number, the setting to allow incoming calls by normal operation without sending a voice message is optional. You may make it possible to perform for every area. Such a function can also be applied to the telephone device 1 described above.

  When the telephone connection device 4 receives a tone signal from the other party by operating the first push button, the telephone connection device 4 sends a ringing signal to the telephone 5 connected to this device to make a ringing tone, and the user performs the telephone 5 by an off-hook operation. When I pick up the handset, the announcement "This party is not a registered number. Please talk carefully while talking. Please press number 1 if you feel suspicious during a call. The voice during the call is recorded through the voice recording means 13 while the call is started. During a call, the voice analysis means 15 monitors the degree of stress of the other party's voice. When the call is terminated by the on-hook operation, one call is made to an emergency contact such as a mobile phone registered in advance in the storage means 12 to notify that an unregistered call has been received. An option that allows the user to select whether or not to notify this one-call transmission is also possible.

  When the telephone analysis device 15 detects stress during the call, the telephone connection device 4 turns on or blinks the monitor lamp 43 to warn the user, and detects the stress or the first push button of the telephone 5 or the outside of the telephone connection device 4. When the switch 44 is pressed, a notification to that effect is sent via e-mail to the e-mail address previously registered in the storage means 12 via the communication network 3. At the same time, a three-party call is set, a call is made to an emergency contact registered in advance in the storage means 12, and a conversation is made through the three-party call. Thereby, crime prevention can be aimed at. Three-way calling may be an option.

  The telephone connection device 4 stores the recording data of the unregistered person based on a preset period after the call ends, and the recording data when there is no problem is about 1 to 3 months (can be set). to erase. When there is a problem such as when stress is detected during a call or when the button of the telephone 5 or the external switch 44 is pressed, the recorded data is stored in a memory card inserted into the memory slot 45 via the external input / output means 17. Since it can be saved, it can also be submitted to the police. It is desirable to prevent the user from easily erasing the recorded data when there is a problem.

  When the telephone connection device 4 detects stress or the first push button of the telephone 5 or the external switch 44 of the telephone connection device 4 is pressed, the telephone connection device 4 is registered in the rejection list of the storage means 12 in association with the recording data, If there is an appointment from the same telephone number again, refer to the refusal list to change the blinking frequency of the warning monitor lamp 43 or use a level meter, 7-segment LED, LCD, etc. You may make it understand the degree.

  Each means that functions in the telephone device of the present invention can be realized by software processing by installing an application program in a computer system such as a mobile phone or a PC, so that a telephone is provided between a commercially available telephone and a telephone line. In addition to the configuration to be inserted as a connection device (system BOX), it can be incorporated into a mobile phone or a CTI server, and can be realized in various forms.

  The embodiments of the present invention have been described above with reference to the drawings. However, the present invention is not limited to the above-described embodiments, and various modifications can be made without departing from the technical scope of the present invention. Is possible.

1 Telephone device 2 Telephone line (PSTN: Public Switched Telephone Network)
3 Communication network (Internet)
DESCRIPTION OF SYMBOLS 4 Telephone connection apparatus 5 Telephone 11 Control means 12 Storage means 13 Voice recording means 14 Voice reproduction means 15 Voice analysis means 16 Input / output means 17 External input / output means 21 Line connection control means 31 Mail transmission means 41 Telephone control means 42 External interface 43 Monitor lamp 44 External switch 45 Memory slot

Claims (6)

Line connection control means for controlling call connection with a telephone line;
Storage means for storing at least audio information as digital data;
Audio reproduction means for analog-converting audio data stored in the storage means for reproduction and output;
Voice recording means for converting voice during a call into digital data and storing it in the storage means;
Voice analysis that determines the stress level of the caller by extracting one or more pieces of information of loudness, frequency, speed, pitch, formant from the voice during the call and comparing it with a predetermined memory pattern Means,
Input / output means for causing each means to function based on a user operation and outputting the result; and
A telephone device comprising: control means for managing cooperation and control of each means.   The voice analysis means performs a Fourier transform on the voice of the other party during a call to extract a frequency power spectrum, and the sound level, frequency, speed, pitch, and formant are selected from the extracted voice component. The degree of stress is determined on the basis of a change in information at a predetermined time, and it is determined that stress is detected when a change in a corresponding component exceeds a predetermined change amount and change frequency set in advance. Item 2. The telephone device according to Item 1.   3. The control unit according to claim 2, wherein when the voice analysis unit detects stress, the control unit calls a telephone number registered in advance by the line connection control unit to perform a three-party call. Telephone device.   3. The apparatus according to claim 2, further comprising: a mail transmission unit that, when the voice analysis unit detects stress, notifies the electronic mail that the stress has been detected via a communication network. The telephone device described in 1.   2. The telephone device according to claim 1, further comprising external input / output means for recording voice data stored in the storage means in a memory card connected externally. A control method in a telephone device having an automatic response and a voice analysis function,
Determining whether the destination phone number is a pre-registered phone number received without ringing,
Sending a voice message prompting the other party to answer whether or not recording is possible for an unregistered phone number;
When a recording approval response is detected, a ringing sound is started to start the call and the voice information during the call is stored as digital data.
Stress is determined by extracting the information of one or more of the loudness, frequency, speed, pitch, and formant from the voice during the call and comparing it with a predetermined memory pattern to determine the stress level of the other party Detecting the stage,
In the case of detecting stress, it has a step of notifying that the stress has been detected to a pre-registered e-mail address by e-mail and calling a pre-registered telephone number to make a three-party call A method for controlling a telephone device.

Images