JPS62285558A - Automatic answering telephone system and individual voice frequency signal detecting means in its system - Google Patents

Abstract

PURPOSE:To prevent a caller from being kept long uselessly by using a recorder capable of random access, supervising a specific identification signal during the transmission of a reply voice so as to attain the changeover to the reply or recording, thereby attaining the reply or recording desired by the caller. CONSTITUTION:The recorder capable of random access such as a disk device is used as the recorder for reply and received voices and a signal detector (individual voice frequency signal detection means) capable of detecting a spe cific identification signal in a received voice signal during the sending of the reply voice at all times is provided. Then a signal representing the detection of the specific identification signal from the signal detector is inputted to control the recorder by a controller. That is, the signal detector 19 detects the sound signal in a specific frequency decided as the specific identification signal in the received signal and applies individual reply/reception corresponding to the identification signal immediately when the specific identification signal is detected.

Description

[Detailed Description of the Invention] 3. Detailed Description of the Invention [Field of Industrial Application] The present invention relates to an answering machine and an individual audio frequency signal detection means used therein, and particularly to an answering machine and an individual audio frequency signal detection means used therein, and in particular to an identification signal from a caller. The present invention relates to an answering machine suitable for selecting a recording area on a recording medium divided by callee, caller, etc., and an individual audio frequency signal detecting means used therein.

[Conventional technology]

When a conventional answering machine receives a call, it reproduces and transmits a prerecorded answering voice, and after that, if necessary, records the message from the caller.

Furthermore, as an improvement plan for this, for example, JP-A-59-1946
Like the answering machine described in Publication No. 1, in addition to the common response voice, a response voice for each called party is prepared, and after the transmission of the common response voice is finished, the calling party can identify the caller. A device has been devised that receives an identification signal and sends out an individual response voice corresponding to the identification signal.

[Problem that the invention seeks to solve]

Of the conventional examples mentioned above, the former always sends out a constant response voice. For example, even when there are multiple called parties to one telephone set, such as in the home, the response voice is unique to each called party. could not be sent.

Further, according to the above-mentioned improvement plan, even an experienced caller has to wait to input the identification signal until the transmission of the common response voice is finished, leading to waste of time and call charges. Furthermore, even if some of the plurality of called parties are absent and others are present, either the common and individual response voices are sent out as if all the called parties were absent, or the present person answers the telephone and the called party I had to reply that the person was absent.

The purpose of the present invention is to perform individual response and reception operations corresponding to the identification signal immediately upon detecting a specific identification signal from the calling party without making the calling party wait unnecessarily. An answering machine that performs individual response and reception operations corresponding to the called party only when the called party corresponding to the identification signal is absent, and has the function of calling the called party by voice if the called party is present. It is an object of the present invention to provide a device and an individual audio frequency signal detection means for detecting a specific identification signal sent from a caller in the answering machine.

[Means for solving problems]

The answering machine according to the present invention uses a randomly accessible recording device such as a disk device as a recording device for the response voice and injured voice, and also transmits a specific identification signal in the received voice signal to the response voice transmitter. A detectable signal detector (individual audio frequency signal detection means) is provided. A control device is provided which inputs a signal indicating detection of the specific identification signal from the signal detector and controls the recording device based on the signal. As described above, as soon as a specific identification signal from a caller is detected, individual response and reception operations corresponding to the identification signal can be performed.

Also, a group of switches for each called party that sets the presence/absence of the called party, a D/A converter that converts the digital signal output from the recording device into an analog audio signal, and the analog audio signal. It is equipped with an amplifier and a speaker that amplifies and converts it into audio. It also includes a connection control circuit for switching the connection between the telephone and the output of the D/A converter to the telephone line. The control device receives the setting states of the switch group, and controls the connection control circuit and the recording device based on the settings.

As described above, only when the called party corresponding to the identification signal is absent, individual response and reception operations corresponding to the called party are performed,
If the called party exists, the called party can be called by voice.

The individual audio frequency signal detection means used in the answering machine may detect a sine wave signal of a single specific frequency or a combination of multiple sine wave signals of specific frequencies such as a PR defect signal as a specific identification signal from the caller. When used, a filter means for removing the specific frequency component from the signal sent from the answering machine to the telephone line is provided, a means for detecting the specific frequency component is provided in the signal path for receiving the signal from the telephone line, and the transmitting, 2 signal paths for reception - 2 wires - 4
It is obtained by connecting to a line via line conversion means.

It is also possible to provide means for extracting the fundamental frequency from the input signal of the filter means, and to remove the specific frequency only when the fundamental frequency becomes 17N (N is an integer) of the specific frequency.

When using a voice signal as a specific identification signal from a caller, a comb filter means for removing harmonic frequency components and a voice recognition means are provided in the signal path for receiving the signal from the telephone line, and the signal transmission path means for extracting the fundamental frequency of the signal, and the frequency characteristics of the filter means are changed based on the fundamental frequency, and the two signal paths are divided into two.
This can also be achieved by connecting to a line via a line-to-four line conversion means.

[Effect]

When an answering call is received, the control device controls the recording device and sends a response voice signal onto the telephone line.

At the same time, the signal detector is operated. Upon detecting the specific identification signal in the received audio signal on the telephone line, the control device immediately stops sending out the response audio signal from the recording device, and detects the response signal on the recording device corresponding to the specific identification signal. A voice signal for response is sent out, and if necessary, a voice signal received from the caller (hereinafter sometimes simply referred to as a reception signal) is recorded in the recording device.

In addition, if the called party is automatically answered and received only when the called party is absent, and the called party is called by voice when the called party is present, then when a call is received, the control device first sends a response voice signal onto the telephone line. do. During this period, when the specific identification signal is detected, the control device checks the status of the callee presence/absence switch group and determines if the called party corresponding to the specific identification signal is present. immediately stops sending out the response audio signal, outputs audio data for calling the called party from the recording device, and outputs the audio data from the D/A converter, amplifier, and speaker. Thereafter, the connection control circuit is controlled to switch the connection of the telephone line from the output of the D/A converter to the telephone set. If the called party is absent, the control device, as soon as it detects the specific identification signal, performs the response and reception recording operations corresponding to the specific identification signal as described above.

Furthermore, the individual audio frequency signal detection means is as follows.

The filter means for removing specific frequency components removes specific frequency components included in the transmitted signal. As a result, a signal with only a small portion of the band of the transmitted signal is transmitted to the line, but this does not affect the caller's listening identification. Furthermore, the signal that bypasses the hybrid circuit and enters the reception path does not include any specific frequency components. Therefore, only the specific frequency for remote control transmitted by the caller reaches the specific signal detection means, so that malfunctions will not occur.

When the transmitted signal is audio, constantly removing a plurality of specific frequency bands causes problems in sound quality.

Speech has a line spectral structure with fundamental frequency intervals.

Therefore, if the fundamental frequency is l/N of the specific frequency, an audio signal component also exists at the specific frequency. The fundamental frequency extracting means extracts the fundamental frequency of the audio signal, and only when this is the specific frequency of 17H, the filtering device removes the specific frequency component from the audio signal. This results in
A voice signal with a portion of the band cut out for a portion of the time is transmitted over the telephone line, allowing the caller to listen to extremely high quality voice. Furthermore, the signal that bypasses the hybrid circuit and enters the reception path does not include any specific frequency components. Therefore, only the specific frequency for remote control transmitted by the caller reaches the specific signal detection means, so that malfunctions will not occur.

The transmitted voice signal is transmitted directly to the line and is listened to by the calling party. The output audio signal bypassing the hybrid circuit is removed by a comb filter means controlled by the fundamental frequency extracted by the fundamental frequency extraction means. Therefore, only the remote control voice signal uttered by the caller is input to the voice recognition means, and no malfunction occurs.

〔Example〕

Hereinafter, one embodiment of the answering machine according to the present invention □ will be described as a first embodiment.
This will be explained using a diagram.

In Fig. 1, the operation panel 1 is a group of switches for setting the basic functions of this telephone answering machine and setting the presence/absence of each called party. There are five switching settings: recording of a call/automatic response based on the presence/absence of the called party, ■recording of a normal call, ■normal call (no recording), and ■playback of recorded audio while the callee is away. This setting state is input to the CPU 4 via the first interface circuit 2 and the pass line 3.

The connection control circuit 6 disconnects and disconnects the telephone 8 and the coupling circuit 9 to the telephone line 7, and is a telephone NCU or the like.
It can be controlled via.

The coupling circuit 9 combines and separates the transmitting signal and the receiving signal on the telephone line, and is a hybrid transformer for a telephone or the like. The relay switch 11 switches the connection between the coupling circuit 9 and the amplifier (13-1) to the D/A converter (12-1), and connects the coupling circuit 9 and the amplifier (13-1) to the D/A converter (12-1) by cpv4 via the third interface circuit 10. It can be controlled. The D/A converter (12-1) converts a digital signal into an analog audio signal. Also, the amplifier (13
-1) amplifies the analog audio signal and outputs the speaker (24).
-1). The first memory 14 is a memory for temporarily recording programs and data, and is a general-purpose RAM or the like. This is used as a work area when the CPU 4 performs arithmetic operations and as a recording area for temporarily used programs. The second memory 15 records data semi-permanently, and is a general-purpose memory such as R01,1. This is used as a recording area for the basic program of the CPU 4, data 3, and a digital voice signal for calling the called party. The ringing signal detection circuit 16 detects a ringing signal on a telephone line when an incoming call is received, and is a telephone tone ringer or the like. When this detects a ringing signal, it sends a signal indicating the detection to the CPU via the fourth interface circuit 17.
Output to interrupt input terminal 4. The signal detection circuit 19 detects a voice signal having a specific frequency and a certain level or more in a received signal, and is a tone detector for telephones or the like. The band pass filter 18 is a filter that selectively passes only the audio signal in the frequency range to be detected by the signal detector 17 in the received voice signal, and the received voice signal is input to the signal detection circuit 19 through this filter. Further, when the signal detection circuit 19 detects an audio signal of a certain level or higher at the specific frequency, it outputs a signal indicating the detection to the interrupt input terminal of the CPU 4 via the fifth interface circuit, and Each associated number is output as a digital signal to the path line 3. The disk device 22 converts the response audio signal and the received audio signal into digital signals and records and stores them, and is a randomly accessible floppy disk device, hard disk device, or the like. If necessary, programs and data necessary for the operation of the CPU 4 can be recorded therein. The disk controller 21 controls reading and writing of the disk device 22, and is an FDC (floppy disk controller) LSI, HDC (hard disk controller) LSI, or the like. This is connected to pass line 3 and C
Read and write operations are performed in response to requests from the PU4. The disk controller 21 sends the response audio signal read from the disk device n to a D/A converter (12-1).
) and input the received audio signal to be written to the disk device n from the A/D converter n. The A/D converter n converts an analog audio signal into a digital signal, and converts the received audio signal into a digital signal.

In addition, the pass line 3 includes a second D/A converter (12-2
), second amplifier (13-2), second speaker (24
-2) Connect.

First, the operation when the basic function setting switch on the operation panel 1 is set to ① automatic response recording when the user is away will be explained with reference to the flowchart of FIG. 2 showing the operation of the CPU 4. First, the CPU 4 controls the connection control circuit 6 via the second interface circuit 5 to open the telephone line 7, and sets the relay switch n to the A side via the third interface circuit 10 to receive an incoming call. Wait (more than 5-
1).

When there is an incoming call, the ring signal detection circuit 16 detects the ring signal on the telephone line 7, and outputs a signal indicating detection to the interrupt input terminal of the CPU 4 via the fourth interface circuit 17 (see 5-2 above). ). In response to this, the cpU 4 controls the connection control circuit 6 to connect the telephone line 7 and the coupling circuit 9 (5-3).

and the disk controller 21 via the path line 3.
, the name of the common response audio file on the disk device 22 and a read instruction signal are output, and the disk controller 21 controls the disk device 4 based on this, and sequentially reads out the digital audio signals recorded in the file. Output to D/A converter (12-1). The digital audio signal is converted into an analog audio signal by a D/A converter (12-1). As described above, the common response voice signal is sent onto the telephone line 7 (5-3). At the same time, the CPU 4 operates the signal detection circuit 19. While sending out the response audio signal, the signal detection circuit 19 detects an audio signal of a specific frequency determined as a specific identification signal in the received audio signal, and sends a signal indicating the detection to the CPU 4 via the fifth interface circuit. When outputting to the interrupt input terminal (
5-5), the cpv4 immediately outputs a reading stop instruction signal to the disk controller 21, and stops reading and sending out the common response audio signal (5-6). Then, a digital signal representing a number corresponding to the audio signal of the specific frequency outputted from the signal detection circuit 19 is inputted via the fifth interface circuit and the pass line 3,
The individual response audio file name and read instruction signal corresponding to the number are output to the disk controller 21. As a result, the individual response voice signal is sent out onto the telephone line 7 in exactly the same manner as the common response voice signal (see 5-7 above). After this transmission is completed, an operation is started to record the received voice signal such as business matters from the calling party in the area on the disk device n that is associated with the number.

That is, the CPU 4 outputs to the disk controller 21 a recording file name corresponding to the number and a write instruction signal. Based on this, the disk controller 21 sequentially inputs the received audio signal converted into a digital audio signal by the A/D converter n, and sends the received audio signal to the disk device n.
Write in the predetermined recording area above (step 5-8 above). The average time required for a disk device to stop reading or writing one file and starting reading or writing another file, that is, from (S-6) to (5-8) in Figure 2. For example, the average time required for a 5-inch floppy disk device is approximately 200 (711
5) is sufficient. When the caller hangs up, the telephone exchange outputs a busy tone signal on the telephone line after a few seconds. When the signal detection circuit 19 detects a busy tone signal in the received audio signal,
A signal indicating the detection should be output to the CPU 4.
In response to this, the PU 4 stops recording the received audio signal, controls the connection control circuit 6, and opens the telephone line 7 (the above S
-10). During the transmission of the common response voice, if a specific frequency signal is not detected in the received voice signal (S-
5), CP and 4 complete the transmission of the common response voice (5-1
1) The received voice signal, such as the caller's business, is transferred to the disk device 2.
Start the operation of recording in the common recording area on 2 (S
-12). The disconnection of the caller's line is detected in exactly the same manner as described above (5-13), recording of the received voice signal is stopped, and the telephone line 7 is released (S-14).

Here, as the audio signal of the specific frequency, for example, a PE (button button) tone for a button telephone is used. This will be explained using a table shown in FIG. This is 0 to 9
and r*J, r◆'' are associated with audio signals of one frequency each for the low group and the high group. Therefore, for example, if you press the button "1", 697 (Hz) and 12
091'-#zJ audio signals are simultaneously sent out onto the telephone line by the tone generator. When a signal detection circuit 19 such as a tone detector detects a PR tone, it outputs a signal indicating detection and also outputs numbers corresponding to the two types of frequencies as digital signals.

As a specific example of the operation described above, a case where the main telephone answering machine is shared by a family will be briefly described. When I receive a call while I'm away, the first thing I do is say, ``I'm away at the moment.

A common response voice signal such as "..." is sent out. When the caller has something to do with his father and presses the pushbutton with the number "1" corresponding to his father, which he had previously learned from his father, the signal detector 19 detects this and immediately sends out a voice signal for common response. The call is canceled and the response voice signal that the father has spoken in advance, such as ``I'm on a business trip to OO.Call number XX,'' is sent out. Next, when the caller speaks his business to his father, this received voice signal is recorded in a recording area (file) for the father on the disk device 22. If a PB tone is not detected while the common response voice signal is being sent, or if it is detected but the number does not correspond to a family member, the caller The received audio signals such as business matters are recorded in a recording area common to the family on the disk device 22.

Further, according to the present invention, when the transmission of individual response voice signals is interrupted and the caller selects a plurality of response voices by number,
Also, when requesting a caller to answer by number, immediately stop sending the answering audio signal that has been sent up to that point as soon as an audio signal of the frequency corresponding to the number is detected,
Transmission of the selected response audio signal or the next response audio signal can be started. For example, if the caller has something to do with the store manager, and you want to differentiate the response voice between transactions and other matters, the response voice signal will be sent as soon as the number corresponding to the transaction is detected. switches to transaction-related matters. Also, when accepting reservations, for example, please press the [Hinichi] button. ”
When requesting a number by sending a response voice signal,
If a caller who is used to the caller presses the button halfway through without finishing hearing the entire message, the caller will immediately add this number and send the next response voice signal, ``Number of people...''. Please press...'' etc. The number changes.

Next, set the basic function setting switch on operation panel 1 to
The operation when the call/automatic response record is set due to absence will be explained with reference to the flowcharts of FIG. 2 and FIG. 4 showing the operation of cpv4.

When a telephone call is received, the CPU 4 performs the series of operations from (5-1) to (S-6) in FIG. 2, as described above. The subsequent operation will be explained with reference to the flowchart in FIG. The CPU 4 looks at the setting state of the callee presence/absence switch corresponding to the voice signal of the specific frequency detected in the received voice signal on the operation panel 1 (t-1).
If the called party is present, first the connection control circuit 6 is controlled to connect both the telephone set 8 and the coupling circuit 9 to the telephone line 7 (t-2). Then, the digital signal of the voice calling the called party, which is recorded on the second memory 15, is sent to the second D/A converter (12) via the primary path line 3.
-2), and the digital signal is converted into an analog audio signal, amplified by a second amplifier (13-2), and emitted by a second speaker (24-2). During this time, the CPU 4 sends a response audio signal on the disk device 22 indicating that the called party is being called (t-3).

When the called party picks up the receiver and the telephone enters the off-footer state, the connection control circuit 6 outputs a signal indicating the off-footer (t-4), and after a while the CPU 4 outputs a signal indicating the answering voice signal during the call. Transmission is stopped (t-5). Furthermore, CP
U4 activates timer 4 when the called party is called, and if the telephone 8 does not become off-footer within a certain period of time (t-6), the callee is absent as shown below in (S-7) in Figure 2. Performs automatic response recording operation. This will also be explained using a specific example of a case where it is used by a family. When a telephone call is received, first a response voice signal such as "Yes, this is 00" is sent out. When the caller has something to do with his father and presses the "1" button corresponding to his father during this time, the signal detection circuit 19 detects the voice signal corresponding to "1", and the CPU 4 detects the voice signal on the operation panel 1. , look at the father's presence/absence switch settings,
If the father is present, a voice is uttered to call the father, such as "Call your father." If no voice signal corresponding to the called party is detected, the cpv4 will emit a ringing voice such as "Telephone." for a certain period of time, and during this time no one will answer the call and the telephone will be in an off-footer state. If not, move on to automatic response recording when you are away.

If necessary, two types of audio signals with specific frequencies may be set for each called party, one for urgent matters and one for other matters, and only when an audio signal corresponding to an emergency is detected, the called party's voice will call. With the configuration shown in FIG. 1, automatic response recording can be easily performed when a corresponding audio signal is detected.

In this embodiment, an example in which a disk device is used as the recording device has been described, but the same effect can be obtained by using a recording device that can be randomly accessed, such as a magnetic drum or a semiconductor memory. Of course. Further, although an example has been described in which an audio signal of a specific frequency, particularly a PB tone signal of a telephone, is used as a specific identification signal, it is of course possible to obtain the same effect by using other specific audio signals. Furthermore, in this embodiment, the case where voice is uttered as a means of calling the called party has been described, but the same effect can be obtained by using other calling means such as lighting a lamp corresponding to the called party. Of course.

Next, an embodiment of the individual audio frequency signal detecting means according to the present invention will be described with reference to diagram m5. FIG. 5 shows an answering machine that can be remotely controlled by a push button dial signal (PR signal) using the present invention.

In FIG. 5, L1*Lt is a telephone line, 31 is a calling signal detection circuit, 32 is a switch, A is a switch drive circuit, Tool is a hybrid circuit, A is a PE signal removal circuit, 36, 3
7, 38, and 39 are selection switches, 40 is a response recording mechanism, 41 is a PE signal detection circuit, 42 is a business recording mechanism, 3 is a control circuit, ■ is a microphone, and 45 is a speaker.

A calling signal is sent from the calling party via the exchange to telephone port N.
When the call arrives at the answering machine of the called party via jJ L, +Lt, the calling signal detection circuit 31 detects this signal and notifies the control circuit 4 of the incoming call. The control circuit 43 closes the switch 32 via the switch drive circuit to close the direct current loop of the telephone line and starts the response recording mechanism 40. At this time, the selection switch 36 activates the response recording mechanism 40.
is selected, and the response message S1 inputted from the microphone 44 selected in advance with the selection switch and recorded in the response recording mechanism 40 is - PE signal removal circuit 35
The signal is transmitted to the calling party through a hybrid circuit that performs 2-wire to 4-wire conversion, and then through the telephone line L1+4.

After the caller listens to the response message S1 and confirms that he is not home, he speaks the message P8, which is transmitted through the telephone line, through the hybrid circuit 34, and recorded as the message selected by the selection switch 37. It is recorded in the mechanism 42. The above-mentioned operation 1b is the same as the known answering machine except that the PE signal band component is removed from the response message St.

A case will be explained in which the caller is the owner of the telephone answering machine, calls the machine from a push-button telephone, and listens to an incoming business message by remote control. The operation from when the owner calls the device until the device answers is the same as described above.

The owner listens to the response message, confirms that the device has responded, and immediately proceeds to the next remote operation. At this time, response messages are still being sent to the telephone line. Remote control begins with a key number that indicates the owner, such as "1".
23" Press the button Press the telephone button "1" Sweat 2" 3
” to send a PE signal to the telephone line. This PE signal reaches the PE signal detection circuit 41 via the vibration circuit A and the vibration circuit A. This key number is determined in advance by the owner and set in the device. The PE signal detection circuit 41 receives the PE signal and transmits the key number "123" to the control circuit. The control circuit 4 identifies the owner by referring to the preset key number and the received key number.

If the call is received from the owner, the control circuit 4 sends the response mechanism 40
and prepare for the next remote operation.

Next, the owner sends a PE signal to the telephone line by pressing the pushbutton "4" of the pushbutton telephone with a number, for example "4", which instructs reproduction of the business message.

This PE signal reaches the PE signal detection circuit 41 via the hybrid circuit A. This number is also set in advance in this device to indicate the reproduction of the business message. PB(I
! Receives the r-thickness detection circuit 41E signal and transmits the number "4" to the control circuit 43. Upon receiving this, the control circuit 4 identifies that the message is to be reproduced, and switches the selection switch 36.
is switched to the output of the message recording mechanism 42, the message recording mechanism is activated, and the recorded message is played back. The PE signal band is removed from the reproduced message by the PE signal removal circuit 35, and the message is sent to the telephone line via the hybrid circuit, so that the owner can listen to the message. While listening to this business message, the next operation, such as replaying the response message, can be performed remotely using a push button. Message input at this time is performed via the telephone line. The speaker 45 is used to listen to incoming messages when the user is at home.

Next, it will be explained that this device can be remotely controlled using PE signals even while reproducing a response message.

The device uses PE signals as remote control signals. P.E.
According to the standards for signals, this signal is O~9, *,
A predetermined low group frequency f Lt + f Lt + f Ls r f
It is a mixture of two audible frequency sine wave signals, one each of Ls and high group frequency fH+ +fH* rfHH. For example, the information of "0" is the frequency f
This is a mixed signal of sine waves of zt and fHl.

The spectrum of the voice in FIG. 12 is shown. This audio also includes a component equivalent to the frequency of the PE signal. If this voice is directly sent to the telephone line via the hybrid circuit 4, if the balance of the hybrid circuit is lost due to a change in the impedance of the line, the voice is bypassed and input to the PE signal detection circuit 41. That is, the PR detection circuit 41 detects a component equivalent to the frequency of the PE signal in this voice and malfunctions. Conventionally, in order to prevent this malfunction, the operation of the FB signal detection circuit was stopped while the response message was being reproduced.

FIG. 13 shows the amplitude frequency characteristics of the PE signal removal circuit 35. The PE signal removal circuit is fL1 to fL41 fgI"
``This is a filter that removes each frequency component of fHs.

The sound shown in FIG. 12 passes through the PE signal removal circuit 35, so that the frequency band components of the PE signal are removed as shown in FIG. 14. In other words, in this device, the audio sent to the line is as shown in Figure 9814, and the audio bypassed by the hybrid circuit 34 is also the same. That is, the signal detection circuit 41 has
The frequency component of the PE signal included in the voice of the response message or message of business is not inputted, and no malfunction occurs. And even if a message voice is sent over the line,
Accurately responds to remote control by the owner's PE signal transmission.

Furthermore, the sound that the owner listens to does not pose any problem as only a part of the sound band is removed as shown in FIG.

FIG. 6 shows another embodiment according to the present invention. The same reference numerals as in FIG. 5 indicate the same parts. 46 and 47 are selection switches.

FIG. 5 shows how the PE signal removal circuit 35 converts the signal sent to the telephone line from the output signal of the answering mechanism or message recording mechanism 42.
The idea is that the frequency band components corresponding to the PE signal are removed using . On the contrary, FIG. 6 shows that the input signal to the response mechanism 40 or the message recording mechanism 42 is recorded by removing the frequency band component corresponding to the PE signal using the PE signal removal circuit 35, and is then played back to the telephone line. It is to be sent to.

In other words, in FIG. 5, the PE signal removal operation is performed on the reproduced signal, and in FIG. 6, the removal operation is performed on the recorded signal.

In FIG. 6, as in FIG. 5, the signal sent to the line has the frequency component corresponding to the PE signal removed, and due to line impedance fluctuations, the hybrid circuit U is bypassed and the signal is sent to the PE signal. The same applies to the signal transmitted to the signal detection circuit 41. As a result, the PE signal detection circuit 41 does not malfunction due to the sending signal, and accurately detects the PE signal even during message playback.
Remote control is possible using the E signal.

The operation of the answering machine shown in FIG. 6 is the same as that shown in FIG. 5.

FIG. 7 shows another embodiment of the present invention. In Fig. 7, the same symbols as in Fig. m5 indicate the same parts. 48 is an adaptive PE signal removal circuit, and 49 is a fundamental frequency extraction circuit.

In FIGS. 5 and 6, components in the frequency band of the PE signal are always removed from the transmitted signal (voice).

As shown in FIG. 13, seven band components are fixed and C is removed, which causes deterioration in audio quality.

The harmonic components of the voice are not always in the frequency band of the PE signal. Therefore, the frequency of the PE signal (fLx - fL4
．． fH, ~f, ys) band, it is only necessary to remove this band component from the voice. As explained above, the voice has a harmonic structure based on the fundamental frequency (the vocal cord imaging frequency) fo as shown in FIG. From this property, when the fundamental frequency f0 of the voice is 1/y of the PE signal, there is always a component in the PE signal frequency band.

The fundamental frequency extraction circuit 49 extracts the fundamental frequency from the audio signal sent to the telephone line. The adaptive PE signal removal circuit monitors the fundamental frequency f0 and removes the component of that frequency for a period of time whose value corresponds to 1/A' of the frequency of the PE signal. By this operation, frequency band components that cause malfunction of the PE signal detection circuit 41 are removed from the transmitted audio signal.

Other operations are the same as in FIG. 5.

FIG. 8 shows another embodiment of the present invention. In FIG. 8, the same symbols as in FIG. 6 indicate the same parts. 7th prisoner is an adaptive band removal operation for the output signal of the response recording mechanism 40 or the business recording mechanism 42, whereas FIG. It performs a band removal operation.

Other operations are similar to those in FIGS. 6 and 7.

FIG. 9 shows another embodiment of the present invention. In FIG. 9, the same reference numerals as in FIG. 7 indicate the same parts. In FIG. 9, 51 is a basic frequency data storage circuit, and 51 is a selection switch.

In FIGS. 7 and 8, input/output signals of the response recording mechanism 40 or the business recording mechanism 42 are processed almost in real time (20 to 20 minutes).
30R5 time delay) to the fundamental frequency extraction circuit 4.
The idea is to extract the fundamental frequency of the voice using 9 and then adaptively remove the frequency band components of the PR signal based on it. The fundamental frequency can be extracted using a known technique called the correlation method or cepstral method.
Performing this in real time requires a huge amount of computation, which requires high-speed and expensive hardware.

The embodiment shown in FIG. 9 solves this drawback.

The standby time of a device is usually expected to be several thousand times longer than the actual operating time. The idea of this embodiment is to make effective use of this waiting time.

The message is recorded into the response recording mechanism or the business recording mechanism without removing the PB signal at one end. When there is no instruction for recording or playback operations to the response recording mechanism or the business recording mechanism, that is, when the device is in a standby state,
Select the fundamental frequency extraction circuit 49 with the selection switch 51,
Play the recorded message and slowly extract the fundamental frequency over time. The fundamental frequency is extracted by reproducing a message for a certain period of time, temporarily storing the data for that period, and performing software processing on the data using, for example, an inexpensive microprocessor over time. By repeating this process, fundamental frequencies corresponding to the messages are sequentially extracted and stored in the fundamental frequency data storage circuit.

Then, there is a reproduction instruction to the response recording mechanism or the business recording mechanism, and when the reproduced audio signal is sent out, the fundamental frequency data corresponding to the reproduced audio signal from the previously stored fundamental frequency data storage circuit 50 is outputted, The frequency band component of the pn signal is adaptively removed from the reproduced audio signal using an adaptive PE signal removal circuit.

The operation of the pond is similar to that shown in FIG.

Audio signals include unvoiced sounds and voiced sounds, and voiced sounds have a fundamental frequency, but unvoiced sounds do not.

Therefore, the adaptive PE signal removal circuit M48 needs to remove all the frequency bands of the PI3 signal for unvoiced sounds as in the embodiment of FIG. The adaptive PR signal removal circuits shown in FIGS. 7-9 have this function.

All of the above explanations have focused on the PR signal as the remote control signal, but it is clear that the PR signal is not limited to this, and for example, a sine wave signal with a single frequency may be used.

FIG. 1O shows another embodiment of the present invention. In FIG. 10, 52 is a comb filter circuit, and group is a speech recognition circuit. This embodiment is an answering machine that allows remote control using voice signals.

The remote control operation of this device is the same as that shown in FIG. 5, except that the remote control signal is a voice, so the explanation will be omitted.

Suppose now that a business message voice is being sent to the line, and while listening to it, the caller utters a voice indicating the next control, such as the word "stop," in order to carry out the next remote control. At this time, two voices, the business message and "stop", are simultaneously input to the comb filter circuit 52.

Therefore, if the comb filter 52 is not used, two voices will be input to the voice recognition circuit 53 at the same time, making it impossible to recognize the remote control voice "stop".

FIG. 15 shows a spectrum when two voices are mixed. In the figure, the solid line indicates the spectrum of the control voice, and the broken line indicates the spectrum of the message voice. This is due to the difference in the fundamental frequency of each voice. It can be seen from the figure that all that is required is to remove the component indicated by the broken line.

In this embodiment, this removal is performed using a known comb filtering technique in the frequency domain.

Figure 16 shows the basic frequency of the message.
The amplitude frequency characteristics of the comb filter are shown.

As shown in the figure, a comb filter has rejection bands arranged like a comb at certain frequency intervals. In the figure, this interval corresponds to the fundamental frequency of the message voice. When the mixed voice passes through the double-shaped filter circuit 52 having such characteristics, it becomes as shown in FIG. &117. As shown in the figure, almost all of the desired control sound can be extracted. Therefore, only the control voice is input to the voice recognition circuit 53 and can be recognized correctly. In other words, remote control operations using control voice are possible even while the business message is being played back.

FIG. 11 shows another embodiment of the present invention. In FIG. 11, the same symbols as in FIGS. 9 and 10 indicate the same parts.

As explained above, the idea of FIG. 10 is to extract the fundamental frequency of the transmitted voice almost in real time, but FIG. 11 has the same idea as the embodiment of FIG. The sound is recorded in the recording mechanism 42, and the fundamental frequency is slowly extracted by the fundamental frequency extraction circuit 49 using the standby time of the device. The comb filter 52 is controlled based on frequency.

The other operations are the same as those in FIG. 10, so the explanation will be omitted.

In FIGS. 10 and 11, if the output of the comb filter circuit 52 is connected to the input of the response recording mechanism,
It is also possible to re-record the response message from the response recording mechanism 40 while listening to the business message.

In the embodiments shown in FIGS. 5 to 11, the response recording mechanism and message recording mechanism have been described as analog magnetic recording such as a cassette tape, but the present invention is not limited to this, and a digital system using an AD or DA converter and a memory may also be used.

〔Effect of the invention〕

As described above, according to the answering machine according to the present invention, a randomly accessible recording device such as a disk device is used as a recording device for answering voice signals and received voice signals,
In addition, the system that sends the response voice monitors a specific identification signal in the received voice signal, and when the identification signal is detected, it immediately switches to the corresponding response or recording operation, so that the caller can When the caller transmits a specific identification signal, the caller can immediately proceed to the response or recording operation desired by the caller, and the caller is not kept waiting unnecessarily. In addition, if there are multiple called parties on A1 telephone, each called person's presence/response
If the called party specified by the caller with a specific identification signal is present, the called party will receive a telephone call, and if the called party is not present, the call will be called. Since the answer is automatically recorded, it is possible to reduce the trouble of relaying the call to other called parties or responding to the caller's absence.In addition, only when the caller has urgent business, a specific identification signal indicating the emergency is sent. The called party is alerted to the incoming call only when the call is sent, and the response is automatically recorded in other cases.
This eliminates the inconvenience of having to wait for non-urgent calls when the called party is busy.

Furthermore, according to the individual audio frequency signal detection means according to the present invention,
Even if the answering machine is sending a signal (voice) to the line, it can identify remote control signals such as FB damage signals or voice signals, which may unnecessarily close the line and increase charges. Since there are no words left, economic efficiency can be improved.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an embodiment of an answering machine according to the present invention, and FIG. 2 is a block diagram showing the CPU 4 during automatic answer recording.
Figure 3 is an explanatory diagram showing the frequency arrangement of the PB tone of the telephone, Figure 1g4 is the CP when recording a call/automatic answer depending on the presence/absence of the called party.
A flowchart showing the operation of U4, FIGS. 5 to 11 are block diagrams showing an embodiment of the individual audio frequency signal detection means according to the present invention, FIG. 12 is an audio spectrum characteristic diagram, and FIG. 13 is a PE' signal. Figure 14 is the amplitude frequency characteristic diagram of the removal circuit, Figure 14 is the audio spectrum characteristic diagram after passing through the PE signal elimination circuit, Figure 15 is the spectrum characteristic diagram when two voices are mixed, Figure 16 is the comb filter circuit diagram. The amplitude frequency characteristic diagram, FIG. 17, is an audio spectrum characteristic diagram after the audio in FIG. 15 has passed through the comb filter circuit. Explanation of symbols 1... Operation panel 2... First interface circuit 3... Pass line 4... CPU 5... Second interface circuit 6... Connection control circuit 7... Telephone line 8...
Telephone 9...Coupling circuit 10...Third interface circuit 11...Relay switch 1
2...D/A converter 13...Amplifier 1
4...First memory 15...Second memory 1
6... Ringing signal detection circuit 17... Fourth interface circuit 18... Band pass filter 19... Signal detection circuit 20... Fifth interface circuit 21... Disk controller 22 ...Disk device Yu...A/D converter U
...Speaker 25...Timer...Hybrid circuit 3'5...PE signal removal circuit 41.
...PR signal detection circuit 48...Adaptive PR signal removal circuit 49...Fundamental frequency extraction circuit (supply)...Fundamental frequency data storage circuit 52...Comb filter circuit Fig. 3 Fig. 4 Fig. 11 Figure 15 n and number n planting

Claims (1)

[Claims] 1. Automatically answers an incoming call, reproduces a response audio signal pre-recorded in a memory from the memory and transmits it to the caller, and In an answering machine that records in a memory an audio signal that conveys the business of a caller, the memory may have prerecorded response voice signals individually for each called party, for each business, or for each called party. There is an individual memory that records messages from the caller side individually, such as by person or task, and can reproduce them, and a memory that is commonly used regardless of the called party or the task. a common memory in which a response voice signal is pre-recorded or a request from a calling party is commonly recorded and can be reproduced; The apparatus includes an individual audio frequency signal detecting means for identifying and detecting a corresponding audio frequency signal when it is sent from the calling party, and a control device, and when an incoming call is received, the audio frequency signal is reproduced from the common memory. When the individual audio frequency signal detecting means detects an audio frequency signal corresponding to an individual area sent from the calling party while the common response audio signal is being sent to the calling party, When the control device learns of this, it stops sending the common response audio signal to the calling party, and reproduces the response audio signal recorded in the individual area corresponding to the detected audio frequency signal. After the transmission is finished, if necessary, the received audio signal from the calling party is stored in a separate area of the memory corresponding to the detected audio frequency signal. An answering machine characterized by recording information. 2. In the answering machine according to claim 1, the individual area of the individual memory corresponds to each of a plurality of called parties, and presence/absence information of each called party is inputted. When the control device receives an incoming call and is transmitting a common response audio signal reproduced from the common memory to the calling party, the individual audio frequency signal detecting means detects a signal from the calling party. When the transmitted audio frequency signal corresponding to a specific called party is detected, after stopping the transmission of the common response audio signal to the calling party, if the called party is absent, the called party The response audio signal recorded in the individual area corresponding to the called party is played back and sent out, and after the transmission is finished, the received audio signal from the calling party is sent to the called party as necessary. 1. An answering machine, characterized in that the information is recorded in a separate area, and if the called party is present, a voice calling operation is performed for the called party. 3. In an answering machine that includes an audio signal transmitting circuit that transmits an audio frequency signal to a telephone line, and an audio signal receiving circuit that receives an audio frequency signal from the telephone line, the audio signal receiving circuit has a caller. An individual audio signal detection circuit for detecting an individual audio frequency signal from the side is provided, and the audio signal transmission circuit is provided with a frequency band removing means for removing the frequency band of the individual audio frequency signal from the audio frequency signal to be transmitted. Individual audio frequency signal detection means characterized by comprising: 4. The individual audio frequency signal detection means according to claim 3, further comprising means for extracting the fundamental frequency from the audio frequency signal on the audio signal transmitting circuit, and the fundamental frequency is the frequency of the individual audio frequency signal. 1/N (however, N
is an integer), the individual audio frequency signal detecting means operates the frequency band removing means. 5. In the individual audio frequency signal detecting means as set forth in claim 4, a storage means for storing the fundamental frequency extracted by the fundamental frequency extracting means is provided, and the individual audio frequency signal detecting operation is not performed. individual audio frequency signal detection means, wherein the fundamental frequency of the audio frequency signal to be sent onto the audio signal transmission circuit is extracted by the fundamental frequency extraction means and stored in the storage means. . 6. In the individual audio frequency signal detecting means according to claim 3, a means for extracting the fundamental frequency from the signal on the audio signal transmitting circuit is provided, and the audio signal receiving circuit is provided with a means for extracting the fundamental frequency from the signal on the audio signal transmitting circuit. Individual audio frequency signal detection means comprising: a comb filter controlled by the extracted fundamental frequency; and speech recognition means to which the output from the filter is input. 7. In the individual audio frequency signal detecting means according to claim 6, a storage means for storing the fundamental frequency extracted by the fundamental frequency extracting means is provided, and the individual audio frequency signal detecting operation is not performed. individual audio frequency signal detection means, wherein the fundamental frequency of the audio frequency signal to be sent onto the audio signal transmission circuit is extracted by the fundamental frequency extraction means and stored in the storage means. .