JPH10313357A - Automatic answering telephone set - Google Patents

Abstract

PROBLEM TO BE SOLVED: To attain remote control by voice and to reproduce a message from a specific talker with priority. SOLUTION: A recognition information storage section 34 stores information to specify a talker and information to specify an action instructed by the user. A voice recognition section 23 references the recognition information storage section to specify the talker, based on voice data received from a voice input section 21. Furthermore, in the case that, a specified talker is recognized by a recognition section 25 to be the user, the voice recognition section specifies an action instruction received from the voice input section by referencing the recognition information storage section to allow an action execution section 26 to execute the action. Thus, the user conducts remote control by voice.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to an answering machine.

[0002]

2. Description of the Related Art An answering machine is widely used to record a message from a calling person (hereinafter referred to as a "speaker") when a user (owner) of the answering machine is absent. I have.

A conventional answering machine outputs a response message and then records a message from the speaker. As the response message, not only a message to be input by the user, but also a plurality of response messages are prepared in advance.
Some are designed to be used by selecting one.

Further, a conventional answering machine can be remotely operated by a tone signal so that a user can listen to a message recorded on the answering machine from a destination.

[0005] Further, there is a type in which a message function by a limited number of members can be realized by using a predetermined personal identification number (tone signal).

[0006]

A first problem is that remote operation cannot be performed from a dial telephone.

The reason is that a remote operation is performed using a tone signal.

[0008] The second problem is that messages from a particular speaker cannot be heard preferentially (or selectively).

The reason is that recording is performed in the order of incoming calls without recognizing the speaker, so that reproduction is possible only in the order of recording.

[0010] A third problem is that it is difficult for the user to memorize the remote operation method.

The reason is that, in order to realize various operations using a limited number of tone signals (dial buttons), the combination of tone signals becomes complicated.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a telephone answering machine in which a speaker can be identified by voice to facilitate remote operation and that a message from a specific speaker can be preferentially reproduced.

[0013]

An answering machine according to the present invention comprises: voice recognition means for performing speaker recognition and voice recognition; and recognition information referred to when the voice recognition means performs the speaker recognition and the voice recognition. Is provided, and action execution means for executing a predetermined action based on the result of the voice recognition is provided, and the speaker is recognized as a specific person by the speaker recognition by the voice recognition unit. In this case, the action executing means executes the predetermined action.

Also, the answering machine of the present invention has a message storage means for storing a message, and when the speaker is recognized as the first specific person by the speaker recognition by the voice recognition unit, An action executing means retrieves the message from the message storage means as the predetermined action.

Further, the answering machine of the present invention, when the speaker is recognized as a person other than the first specific person by the speaker recognition by the voice recognition unit, the action executing means as the predetermined action, The input message is stored in the message storage means.

Alternatively, the answering machine of the present invention comprises: voice recognition means for performing speaker recognition; recognition information storage means for storing recognition information to be referred to when the voice recognition means performs the speaker recognition; Action executing means for executing an action, wherein when the speaker is recognized as a specific person by speaker recognition by the voice recognition unit, the action executing means executes the predetermined action. It is characterized by.

Also, the answering machine of the present invention has an action storage means for storing first personal name information and first action information in association with each other, and a message storage means for storing a message. When the person is the person represented by the first name information, the action execution means executes the first action as the predetermined action, and adds the first name information to the input message. And storing it in the message storage means.

Further, the answering machine of the present invention is characterized in that when the specific person is other than the person represented by the first name information, the action executing means performs the predetermined information as the predetermined action by adding the recognition information to the input message. And the name information of the specific person obtained from the message is added and stored in the message storage means.

Further, the recording medium of the present invention includes a program for performing speaker identification and voice identification based on input speech data with reference to recognition information stored in a storage device, and a program for identifying the speaker. A program for executing a different action is stored.

[0020]

Next, an embodiment of the present invention will be described with reference to the drawings.

FIG. 1 shows an embodiment of the present invention. FIG.
The answering machine includes a telephone body 1, a data processing device 2 operated by program control, a storage device 3 for storing information, and a recording medium 4 on which a program is recorded.

The telephone body 1 has a normal telephone function and an answering function for detecting an incoming call and automatically off-hook.

The storage device 3 includes an action storage unit 31 and
A message storage unit 32, an authentication information storage unit 33, and a recognition information storage unit 34 are provided.

The action storage unit 31 stores in advance response processing for a person who is supposed to make a call, using the name of the person as a key.

The message storage unit 32 stores recorded data for each person who makes a call. Here, time information indicating the time of the call is added to each recording data.

The authentication information storage unit 33 stores in advance information (password) for determining whether or not a speaker is a user when there is a possibility that the speaker is a user.

The recognition information storage unit 34 stores in advance the name of a person who is supposed to make a call, information necessary to identify the person by voice, and information for the user to specify an action. I have.

As the recording medium 4, various recording media such as a magnetic disk and a semiconductor memory can be used.
The recording medium 4 stores programs for realizing the voice input unit 21, the control unit 22, the voice recognition unit 23, the voice synthesis unit 24, the authentication unit 25, the action execution unit 26, and the voice output unit 27. Is realized by the data processing device 2 reading the program from the recording medium 4 and executing the program.

The voice input unit 21 obtains voice data from the telephone body 1 and passes it to the control unit 22.

The voice recognition unit 23 receives the voice data from the control unit 22, determines the speaker by referring to the recognition information storage unit 34, and passes the determination result to the control unit 22.

The voice synthesizing section 24 synthesizes voice from the character string passed from the control section 22 and passes it to the voice output section 27.

The authentication unit 25 refers to the authentication information storage unit 33 and determines whether the password is correct when the result of the speaker identification is determined to be the user.

The action execution unit 26 refers to the action storage unit 31 and retrieves a message to the corresponding speaker. Further, the recording data is stored in the message storage unit 32. Here, information indicating a speaker and time information are added to the recording data.

The voice output unit 27 transmits the data received from the control unit 22 to the telephone body 1.

The control unit 22 passes the data received from each unit to each unit that performs the following processing.

Next, the operation of the answering machine according to this embodiment will be described with reference to FIGS.

When the telephone body 1 detects an incoming call in step A1, it determines in step A2 whether or not the answering machine is set. If the answering setting has been made, the process proceeds to step A3. If the answering setting has not been made, the process ends. However, when the ringing signal is received a predetermined number of times or more, the setting may be automatically changed to the absence setting, and the process may proceed to step A3.

When the telephone body 1 determines that the absence setting has been made in step A2, it automatically goes off-hook in step A3, sends a response message, and inputs a voice to the speaker, for example, a name. Request voice input for In response to this request, the audio signal input by the speaker
It is provided from the telephone body 1 to the voice input unit 21 of the data processing device 2. The voice input unit 21 holds the input voice as voice data (step A3), analyzes the voice data, and passes the analyzed data to the control unit 22. The control unit 22 transmits the data from the voice input unit 21 to the voice recognition unit 23
Pass to.

Upon receiving the data from the control unit 22, the voice recognition unit 23 specifies the speaker by referring to the recognition information storage unit 34 in step A4.

If the speaker cannot be specified, the voice recognition unit 23
Notifies the control unit 22 of this. The control unit 22
Upon receiving the notification that the speaker cannot be identified, the voice input unit 21
Receives the voice data held by the user and passes it to the action execution unit 26 together with the notification of the unknown speaker. Upon receiving the voice data accompanied by the unknown speaker notification, the action executing unit 26 writes the voice data to which the information indicating the unknown speaker and the time information are added to the message storage unit 32 as recording data in step A10.

On the other hand, if the speaker can be specified in step A4, the voice recognition unit 23 further determines in step A5 whether or not the speaker is the specified person. This decision
For example, based on the probability that the speaker is the identified person, if the probability is equal to or greater than a predetermined value, it is correct (almost certainly).
If it is equal to or less than the predetermined value, it is determined that there is an error (there is a possibility of another person). In particular, when the specified person is considered to be a user, it is desirable to increase the criteria for determination.

If it is determined in step A5 that an error has occurred, the speaker is requested to input speech again, and steps A3 and A4 are repeated to specify the speaker. If the person specified this time is the same as the person specified last time, in the second step A5, it is determined that the determination result is correct and the process proceeds to the next step. The voice input request to the speaker is sent to the control unit 22, the voice synthesis unit 24, and the voice output unit 27.
It is performed by

Thus, the voice recognition unit 23 specifies the speaker and notifies the control unit 22 of the result. The control unit 22
If the specified speaker is the user himself, the voice synthesis unit 24 and the voice output unit 27 request input of a password. Then, the input password is input to the voice input unit 2.
1 and passed to the authentication means 25. Authentication means 25
Determines whether the input password is correct with reference to the authentication information storage unit 33 (step A6). If the password is correct, the control unit 22 is notified that the speaker has been confirmed as the user himself.

When it is confirmed that the speaker is the user, the control section 22 requests an action instruction from the voice synthesis section 24 and the voice output section 27. The action instruction by voice from the speaker is transmitted from the voice input unit 21 to the voice recognition unit 23 through the control unit 22. Voice recognition unit 23
Acquires the contents of the specified action with reference to the recognition information storage unit 23, and requests the action execution unit 26 to execute the obtained action via the control unit 22.

The action execution section 26 executes the action requested in step A7. The actions here include, for example, reading out the recorded data stored in the message storage unit, or changing the action stored in the action storage unit.

In step A6, if the input password is incorrect, the process may be terminated as it is, or the password may be re-entered, and the process may be terminated if the result is incorrect. Is also good.

On the other hand, if the specified speaker is a person other than the user, it is determined that the speaker is not the user without requesting the input of a password (step A6). In this case, the control unit 2
2 notifies the action execution unit 26 of the name of the speaker specified by the voice recognition unit 23. The action execution unit 26
In step A8, the action storage unit 31 is searched to determine whether an action is designated for the notified speaker. If an action has been specified, the action execution unit 26 executes the specified action in step A9. This action includes, for example, transmitting a message.

If no action is specified, the action executing section 26 informs the control section 22 of the fact. The control unit 22 receives the notification, and
And sends it to the action execution unit 26. The action execution unit 26 adds the speaker name and the time information to the received voice data, and writes the recording data in the message storage unit 32.

As described above, in the present embodiment, the user can remotely control the apparatus from outside by voice. Specifically, it is possible to reproduce only a message from a specific speaker. In addition, a message can be transmitted to a specific speaker.

It is needless to say that the message stored in the message storage section 32 can be reproduced using the telephone body 1. At that time, of course, the reproduction can be performed for each specific speaker.

In the above embodiment, the voice input unit 2
1 is held as voice data, but before step A10, a response message such as "Please give a message" is transmitted to the speaker, and then transmitted. Step A1
0 may be stored directly in the message storage unit 32.

Further, in the above-described embodiment, only the case where only voice is handled has been described. However, remote operation using tone signals may be enabled.

[0053]

Next, an embodiment of the present invention will be described with reference to FIG.

Referring to FIG. 3, the action storage unit 31
In the message “Message 1
Tell "is specified. The recognition information storage unit 3
4 stores information on the persons A, B, and C. The user is B, and the authentication information storage unit 33 stores a password in advance.

If the answering machine is set to answering machine, and B calls the answering machine from whereabouts, the answering machine setting is determined by the telephone body 1 and the telephone goes to an off-hook state, and the telephone is received by the data processing apparatus. Is conveyed.
B gives his name in response to a voice input request from the answering machine. The voice input unit 21 records this (step A).
3) Analyze and provide the speech data to the speech recognition unit 23.
When the voice recognition unit 23 determines that the speaker is B (steps A4 and A5), the authentication unit 25 requests the input of a password, and determines whether or not the user is himself using the input password. Confirm (step A6). When it is confirmed that the speaker is B, the data processing device 2 makes a request for specifying an action, and executes the specified action (step A7).

[0056]

According to the present invention, since remote control by voice is possible, the operation is simple and the operability is improved.

More specifically, remote operation can be performed from a dial telephone. Also, a message from a specific speaker can be selected and reproduced. Furthermore, there is no need to learn complicated operations.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention.

FIG. 2 is a flowchart for explaining the operation of the answering machine of FIG. 1;

FIG. 3 is a block diagram showing one embodiment of the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Telephone main body 2 Data processing device 3 Storage device 4 Recording medium 31 Action storage unit 32 Message storage unit 33 Authentication information storage unit 34 Recognition information storage unit 21 Voice input unit 22 Control unit 23 Voice recognition unit 24 Voice synthesis unit 25 Authentication unit 26 Action execution unit 27 Voice output unit

Claims (7)

[Claims] 1. A voice recognition unit for performing speaker recognition and voice recognition, a recognition information storage unit for storing recognition information to be referred to when the voice recognition unit performs the speaker recognition and the voice recognition, and the voice Action executing means for executing a predetermined action based on a result of the recognition, and when the speaker is recognized as a specific person by the speaker recognition by the voice recognition unit, the action executing means An answering machine characterized by performing an action. And a message storage unit for storing a message, wherein when the speaker is recognized as a first specific person by speaker recognition by the voice recognition unit, the action execution unit is configured to perform the predetermined action. 2. The answering machine according to claim 1, wherein said message is retrieved from said message storage means as an action. 3. When the speaker is recognized as a person other than the first specific person by the speaker recognition by the voice recognition unit,
3. The answering machine according to claim 2, wherein said action execution means stores the input message as said predetermined action in said message storage means. 4. Speech recognition means for performing speaker recognition, recognition information storage means for storing recognition information to be referred to when the speech recognition means performs the speaker recognition, and action execution means for executing a predetermined action An answering machine, wherein the action execution means executes the predetermined action when the speaker is recognized as a specific person by the speaker recognition by the voice recognition unit. 5. An apparatus comprising: action storage means for storing first personal name information and first action information in association with each other; and message storing means for storing a message, wherein said specific person is said first personal name information. When the person is represented by information, the action execution means executes the first action as the predetermined action, and adds the first person name information to the input message to store the message in the message storage means. The answering machine according to claim 4, wherein said answering machine is stored. 6. When the specific person is other than the person represented by the first personal name information, the action executing means may execute the predetermined action as the predetermined action, wherein the specific person is obtained from the recognition information in an input message. 6. The answering machine according to claim 5, wherein name information is added and stored in said message storage means. 7. A program for performing speaker identification and voice identification based on input speech data with reference to recognition information stored in a storage device, and a program for performing different actions depending on the identified speakers. A recording medium characterized by storing therein.