JPS5812783B2 - Automatic telephone answering recording device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a discrimination circuit that can distinguish a voice signal that can be used in an answering machine, etc. from an intermittent tone such as a busy tone, a continuous tone such as a dial tone, or a no signal state. Patent Application No. 1973-53202 (Patent Publication No. 1973)
-32005).

As is well known, an answering machine has a response tape on which response words are recorded and a reception tape on which a message signal from a customer is recorded.When a call signal from a customer arrives, the answering tape is first recorded. When the response is completed, the receiving tape is started and recording of the customer's message signal begins.

The running of the receiving tape is controlled via a fixed timer circuit or an audio switch circuit.

However, in a manner controlled via a voice switch circuit,
When the customer finishes using the device and puts down the handset, the busy tone or dial tone sent from the central office to the device causes the device including the voice switch circuit to continue to malfunction. There is.

Conventionally, in order to prevent such malfunctions, there is a method of attenuating frequency components corresponding to busy sounds and dial sounds using a filter circuit.

However, this method is prone to malfunctions due to harmonics.
Moreover, it has the disadvantage that the circuit becomes complicated.

In order to improve this drawback, various methods without using a filter circuit have been proposed.

For example, such as the above-mentioned patent application.

However, in the circuit configuration used in this patent application, the oscillation output width of the monostable multivibrator circuit is limited to the signal part of the busy tone (
In the part where the signal is sent out, the width cannot be made smaller than 0.5 seconds), and the intermittent sound with the waveform shown in Figure 3, which is different from the busy sound, is used in some foreign countries. If this voltage is applied, a malfunction will occur (the device cannot be restored to the standby state).

) In the past, the discrimination circuit for discriminating the intermittent sound shown in Fig. 3 was considered difficult, but in the present invention, a retriggerable monostable multivibrator circuit C (hereinafter referred to as a single multivibrator circuit)
By using the , the intermittent sound shown in Figure 3, the intermittent sound that repeats transmission and interruption at 0.5 second intervals like a busy tone, the continuous sound like a dial tone, or even a no-signal state can be detected as a voice signal. To provide a signal discriminating circuit capable of discriminating signals with a simple circuit configuration.

The operation of one application example will be specifically described below with reference to the accompanying drawings.

FIG. 1 is a circuit example showing one embodiment.

Figure 2 is an explanatory diagram showing the operation points for voice signals and busy sounds, and continuous dial sounds and no-signal states can be easily explained in relation to busy sounds, so Figure 2 It is omitted in .

As mentioned above, FIG. 3 shows the intermittent tone used in some foreign telephone offices, and when the customer hangs up the handset, it becomes a call end signal sent from the office to the equipment side.

Now, in Figure 1, 1 is a rectifier and smoothing circuit, 2 is a switching circuit using a Schmitt circuit, 3 is a single multi-circuit integrated circuit, and 4 is a gate circuit, which is the same as 3 because only one IC is required. using a circuit.

Of course, a dedicated gate circuit, such as a NAND circuit and an inverter circuit, may be appropriately combined.

In the embodiment, the SN74123N is used in which these two circuits are housed in the same package.

5 is a timer circuit whose timed operation is controlled by a semi-fixed resistor V
For example, if the customer's speech is not interrupted for more than 7 seconds, the transistors Q4 and Q5 are turned ON while repeatedly charging the integrating capacitor C5 using the customer's voice signal. None, the load relay Y is kept in an operating state, and the running state of the receiving tape is maintained through the contacts of the relay, which is well known and not shown, to record all of the customer's voice signals.

Then, when the customer finishes using the device and places the handset, the device receives an intermittent or continuous sound, or when there is no signal, charging of the capacitor C5 stops, and approximately 7 seconds have passed. Then, the discharge of the charge in C5 ends and Q
4 and Q5 are turned OFF to restore relay Y, and at this time, although not shown, the apparatus is configured to stop running the receiving tape and at the same time restore the apparatus to a standby state.

More specifically, when a customer's voice signal is applied to the input terminal IN, it is voltage-doubled and rectified by diodes D1 and D2, and smoothed by capacitor C2.

Therefore, the DC voltage that is the envelope of the audio signal during the time axis t0 to t11 shown in FIG. 2A is applied to the base of the transistor Q1.

If the operating point of Q1 and Q2 is set to V1 in diagram A, this V
A peak portion of the audio signal that exceeds 1 causes Q1 and Q2 to be inverted.

That is, diagram B shows Q2, and diagram C shows changes in the collector voltage of Q1, and the output of Q1 changes from H to L and the output of Q2 changes from L to H due to the audio signal exceeding the operating point V1.

For example, on the time axis to, when the output changes in this way, the output of the inverter circuit 7 is H because terminal 1 is grounded in the single multi-circuit 3, and from the above, the input of terminal 2 is also H. Therefore, the two input terminals of the 6 AND circuit become H at the same time on the time axis t0.

Now single multi circuit 3 is triggered and capacitor C3
A pulse having a predetermined pulse width determined by the time constant of resistor R5 is oscillated.

In the embodiment, the pulse width was set to about 0.2 seconds, which is the same as that of the gate circuit 4 using a single multi-circuit.

The reason for this will be detailed later.

This oscillation output is taken out from terminal 4.

Therefore, as shown in diagram D, on the time axis to, the terminal 4 (Q output) that was H changes to L.

However, since the single multi-circuit 3 is configured to be retriggerable, it is retriggered at time axis t2 while the pulse of about 0.2 seconds is being oscillated as described above, and the oscillation is It is further extended for about 0.2 seconds, and during this extended pulse oscillation, a re-trigger is applied at time axis t4, and the pulse oscillation is further extended for about 0.2 seconds, and eventually reaches time axis t0 in Figure D. An output as shown at ~t12 is obtained at terminal 4.

This is because the amplitude of the audio signal changes from moment to moment, and depending on the audio signal that lasts for one second on average, switching of Q1 and Q2 can repeat the conduction and non-conduction of the circuit several times, and within 0.2 seconds. This is because re-triggering is easy.

According to experiments, re-triggering occurred even with this pulse width of 0.1 seconds.

Now, as mentioned above, the output shown in diagram D (L between t0 and t12)
Since the output (output) is applied to the terminal 9 of the single multi-circuit 4 and inverted by the inverter circuit 8, the output in figure D becomes as shown in figure E, and is applied to one input terminal of the AND circuit 9, and is inverted by the inverter circuit 8. The terminal is connected via terminal 10 to the collector side of transistor Q1.

Therefore, the single multi-circuit 3 oscillates when the two input terminals of the AND circuit 9 become H at the same time.

The point in time when the voltage becomes H at the same time is clear by comparing diagram C showing the change in the collector voltage of Q1 and diagram E showing the output of the inverter circuit 8.

That is, the time axis t1-t2. t3-t4, t5-t6.
During each period from t7 to t8, from t9 to t10, from t11 to t12, the two input terminals of the AND circuit 9 become H at the same time, and at the same time begin to become H at t1, t3, t5, t7,
A trigger is applied at each time point t9 and t11, and the single multi-circuit 4 oscillates.

Since this circuit can be retriggered, an L output is obtained at the terminal 12 on the time axis 11 to t13 as shown in diagram F.

Therefore, the bias applied through resistor R7 during this time is cut off, so transistor Q3 is turned off.
Charging of capacitor C5 is started via resistor R8 and diode D3, and transistors Q4 and Q5 are ONK driven via semi-fixed resistor VR to operate relay Y as a load.

In reality, the capacitor C5 is charged to a predetermined potential in advance, and the relay Y is connected to the collector side of Q4 and Q5 when the response tape finishes running.

On the other hand, if the customer's voice is interrupted, for example, time axis t1
3 to t14, the output of the terminal 12 becomes H, which turns Q3 ON and stops charging the capacitor C5.

However, if the time during which the audio signal is interrupted is within the aforementioned 7 seconds, relay Y can be continuously maintained in the operating state by the charge of C5.

Since the resistance value of R8 can be selected to be low, when the next audio signal is added after a break in speech, C5 will be charged relatively quickly, and even if the audio signal is added at the last 7 seconds mentioned above, relay Y will be charged. can be reliably maintained in continuous operation.

Next, the operation of the present invention after the customer finishes using the device and puts down the handset will be explained.

First, a case will be described in which the busy sound is applied after time axis t14 shown in Figure A.

The width of the signal portion between t14 and t16 and the width of the no signal portion between t16 and t17 are both 0.5 seconds.

Due to this busy tone, the collector voltages of Q2 and Q1 are
The pulse changes as shown in the figure, and as is clear from the above, the single multi-circuit 3 oscillates a pulse as shown in D at time axis t14.

That is, the period between t14 and t15 in Figure D is the oscillation output of this pulse, and its pulse width is about 0.2 seconds.

This oscillation output is applied to the single multi-circuit 4 via the terminal 9 as described above, but the condition for this circuit to oscillate is when the two input terminals of the AND circuit 9 become H at the same time, as described above.

This point can be understood by comparing diagrams C and E as described above.

However, there is no point in time when the signals become H at the same time (as mentioned above, in order to become H at the same time, it is necessary to re-trigger and extend the pulse width shown in Figure D).

Therefore, the transistor Q3 is kept in the ON state, and the capacitor C5 cannot be charged, and about 7 seconds after the discharge of the electric charge of the capacitor C5 is completed, the transistor Q3 is kept in the ON state.
4 and Q5 are turned OFF to restore relay Y.

It is known that when this relay Y is restored, the apparatus is restored to a standby state and at the same time the running of the receiving tape is stopped.

Next, if the customer puts down the handset and a dial tone is sent, the explanatory diagram between time axis t14 and t16 is appropriate, and when the customer puts down the handset, the input terminal on the device side becomes a no-signal state. In this case, the explanatory diagram between time axes t16 and t17 is appropriate.

That is, in either case, the single multi-circuit 4 (gate circuit) does not oscillate, so 05 cannot be charged, and the relay Y can be restored after approximately 7 seconds have elapsed.

This monostable multi-4 is used to detect the AND output of the waveforms in Figures C and E, so if there are extra NAND circuits and inverter circuits used in other circuits,
A suitable combination of these can also be used in place of the monostable multi-4.

On the other hand, in some foreign stations, when a customer puts down the handset, intermittent tones with alternating amplitudes and signal widths are sent to the equipment, as shown in FIG.

Conventionally, it has been considered difficult to develop a discrimination circuit that can operate reliably for intermittent sounds having different amplitudes and signal portion widths.

Since the minimum width of this signal portion is 0.25 seconds, the oscillation output width of the single multi-circuit 3 was set to the aforementioned 0.2 seconds so as to be smaller than this 0.25 seconds.

This is because if you compare the time axes t14 to t15 in diagrams C and E, it is clear that the oscillation output width of the single multi-circuit 3 is 0.
If it is longer than 25 seconds, the single multi-circuit 4 will oscillate during the 0.35 second no-signal period, making it impossible to restore the relay Y and causing malfunction.

In this way, if the width of the single multi-circuit 30 oscillation output is set below the minimum width of the signal part of the intermittent sound applied to the equipment side, no matter how long the width of the signal part beyond this is, there will be no It has the advantage of being completely unaffected by the width of the signal section.

Further, the operating point V1 of the switching circuit 2 composed of the transistors Q1 and Q2 can be set to either V1' or V'' in FIG.

If it is set to V1'', the signal portion between t20 and t21 operates in the same way as the non-signal portion, so that no malfunction will occur as explained above.

Normally, the operating point V1 is set to the level of V1', but since it is normally input through an amplifier with an ALC circuit, the ALC signal part (t22 to t23) with a large amplitude
It is also conceivable that the circuit works, the low-level signal portion (t20 to t21) is suppressed to an even lower level, and the operating point V1 moves to V1''.

In conventional signal discrimination circuits, such a shift in the operating point often causes malfunctions.

As mentioned above, the oscillation output width of a single multi-circuit that can be retriggered is determined by the signal part (sending part) of the intermittent sound applied to the device side.
If the width is set appropriately so that it is re-triggered by the audio signal, it will work reliably even with intermittent sounds that have a signal part larger than the set width, and the intermittent It is not affected by the amplitude of the sound, and since the main circuit of the present invention is integrated circuit, it is possible to save labor, and the oscillation output width of the monostable multi-3 is 0.1 as mentioned above.
Since the time can be set to between seconds and 0.25 seconds (within a range of 0.1 to 0.5 seconds when used domestically), it has the advantage of facilitating component selection.

[Brief explanation of the drawing]

FIG. 1 is a circuit diagram showing an embodiment of the present invention, FIG. 2 is an explanatory diagram of operation points corresponding to input signals, and FIG. 3 is an intermittent signal employed in some foreign stations.

Claims (1)

[Scope of Claims] 1. In an automatic telephone answering recording device that configures a closed circuit in response to an incoming telephone call, sends out prerecorded response terms, and then records a customer's message on a receiving tape, a switching circuit that is switched by an audible signal (including an audio signal); a first mono multi circuit that is triggered by the output of the switching circuit; and an AND output of the output of the switching circuit and the first mono multi circuit. an electromagnetic circuit including a second mono multi-circuit that is triggered; a timer circuit that continues to operate based on the output of the second mono multi-circuit; and a relay that is connected to pause when the timer circuit stops operating; An automatic telephone answering and recording device comprising: a recovery circuit connected to restore the device by the electromagnetic circuit described above.