JPH07107198A - Telemarketing equipment - Google Patents

Abstract

PURPOSE:To exactly and surely conduct telemarketing, even in the case a speaker of the other party is changed on the way. CONSTITUTION:This equipment is provided with plural filters for filtering different frequency (fa), (fb) components from a signal Sr of a response voice, to put it concretely, band pass filters 3a, 3b, and a control part 4 for sampling the magnitude of filtering signals Sa, Sb obtained from each band pass filter 3a, 3b, and also, discriminating a speaker of the other party at an interval of a prescribed time from a relation of magnitude Pa (Pac), Pb (Pbc) of the filtering signals Sa, Sb in each frequency (fa), (fb) component, and sending out a voice message from the beginning, when it is discriminated that the discriminated speaker is changed.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a telemarketing device for unmanned telemarketing using a telephone line.

[0002]

2. Description of the Related Art Conventionally, the telephone numbers of a plurality of customers are registered in advance, and the dialing time corresponding to each telephone number is set at a predetermined time interval. When the dialing time comes, the corresponding telephone numbers are automatically set. Dialing (calling) by dialing, sending the voice messages stored in the memory to the other party connected to the line in order, and recording the response voice of the other party who responded to this voice message on magnetic tape etc. Telemarketing devices (automatic call / automatic call type communication terminal devices) are known.

[0003]

By the way, when a customer is called by such a telemarketing device, it is not uncommon for the other party, for example, a child who cannot respond to pick up the telephone receiver. In this case, even if the phone changes to an adult who can answer, the adult will hear the voice message from the middle, and it will not be possible to perform accurate telemarketing (service).
Depending on the situation, there was a problem that telemarketing could not be performed at all.

The present invention solves the problems existing in the prior art as described above, and a telemarketing apparatus capable of always performing accurate and reliable telemarketing even when the speaker of the other party changes in the middle. For the purpose of providing.

[0005]

DISCLOSURE OF THE INVENTION The present invention is a telemarketing device adapted to sequentially send voice messages stored in a memory 2 to a party connected to a line by dialing and record a response voice of the party. 1 is configured, in particular, a plurality of filters for filtering different frequency components, for example, a low frequency component fa and a high frequency component fb from the response voice signal Sr, specifically band pass filters 3a and 3b, Filtering signals Sa and S obtained from the bandpass filters 3a and 3b
While sampling the magnitude of b, each frequency f
If the speaker of the other party is identified at predetermined time intervals based on the relationship between the magnitudes Pa (Pac) and Pb (Pbc) of the filtered signals Sa and Sb in the a and fb components, and it is determined that the identified speakers are different. It is characterized by including a control unit 4 for sending a voice message from the beginning.

In this case, the bandpass filters 3a, 3
Analog-digital converters 5a and 5b for converting the filtered signals Sa and Sb obtained from b into digital signals can be provided. In addition, the filtering signals Sa and S
The size of b is the value Ua ..., Ub ...
Can be integrated and obtained as the integrated value or the average value as it is, and the control unit 4 controls the frequencies fa and fb.
The speaker can be identified by the ratio of the sizes of the filtered signals Sa and Sb in the component.

[0007]

According to the telemarketing device 1 of the present invention, first, for example, the low-pass frequency fa component and the high-pass frequency fb component are respectively filtered from the response voice signal Sr by the bandpass filters 3a and 3b, Filtered signals Sa and S at respective frequencies fa and fb
b is obtained. In this case, the analog-digital converter 5
By providing a and 5b, the obtained filtering signals (analog signals) Sa and Sb are converted into digital signals, respectively.

On the other hand, the control unit 4 controls the filtering signals (digital signals) Sa and S at the respective frequencies fa and fb.
From the relationship between the sizes Pa and Pb of b, the speaker of the other party is identified at predetermined time intervals. Specifically, the magnitudes of the filtered signals Sa and Sb are sampled a plurality of times over a certain period of time, and the values Ua ..., Ub ... Obtained by sampling are integrated for each frequency fa, fb component unit, and the integrated value (or Filtered signals Sa, S as (average value)
The magnitudes Pa and Pb of b are obtained, and the ratio of the magnitudes Pa and Pb of the obtained filtering signals Sa and Sb, that is, Pa
The speaker is identified by / Pb.

The control unit 4 also monitors the identified speaker,
The magnitudes of the filtering signals Sa and Sb change, that is,
By changing Pa to Pac and Pb to Pbc,
If it is determined that the speakers are different, the voice message is sent from the beginning. In this case, the voice message stored in the memory 2 is read from the beginning. As a result, even if a child who cannot answer first picks up the handset of the telephone and turns into an adult who can answer in the middle, the adult can listen to the voice message from the beginning and perform accurate and reliable telemarketing. It will be possible.

[0010]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, preferred embodiments according to the present invention will be described in detail with reference to the drawings.

First, in order to clarify the present invention, an overall schematic configuration of the telemarketing apparatus 1 will be described with reference to FIG.

In FIG. 2, reference numeral 20 is a rack, and the rack 20 is provided with a telemarketing device main body 21 connected to a telephone line. Telemarketing device body 21
The front panel 21p of the display 2 for performing various displays
A keyboard 23 for arranging 2 and for inputting
Distribute. Further, a printer 24 for printing out various data and the like is connected to the telemarketing apparatus main body 21, and tape recorders 25, 26, 27 for recording the response voice received from the other party on a magnetic tape are connected. In addition, 28 is a telemarketing apparatus main body 21.
The reference numeral 29 designates a manually usable telephone connected to the tape player, and 29 designates a tape player for reproducing a magnetic tape.

Next, the main body 21 of the telemarketing device.
The configuration of the main part of the present invention provided for will be described with reference to FIG.

In FIG. 1, reference numeral 4 denotes a control unit including a CPU, a ROM, a RAM and the like, which executes various arithmetic processes, control processes and the like. A voice synthesis unit 11 controlled by the control unit 4 converts the voice message read from the memory 2 and the memory 2 using a voice ROM that stores a predetermined voice message as data into an analog signal. It includes a digital-analog converter and the like.

On the other hand, 12 is 2 lines / 4 connected to the telephone line.
This is a line converter, and the voice synthesizer 11 is connected to the input side of the 2-line / 4-line converter 12 and one input unit of the difference unit 13 is connected to the output side thereof, and the 2-line / 4-line converter is also provided. Bowl 1
The input side of 2 is connected to the subtractor 1 via the echo canceller 14.
3 is connected to the other input section. The output unit of the difference unit 13 is connected to the control unit 4 via the first bandpass filter 3a and the analog-digital converter (A / D converter) 5a, and the output unit of the difference unit 13 is further It is connected to the control unit 4 via the two bandpass filter 3b and the A / D converter 5b. In this case, the first bandpass filter 3
The filtering frequency of a is selected as the low frequency fa and the filtering frequency of the second bandpass filter 3b is selected as the high frequency fb.

Next, the operation of the telemarketing device 1 according to the present invention will be described with reference to the flow charts shown in FIGS. 5 to 7 and FIGS.

First, the telephone numbers of a plurality of customers are registered in the telemarketing apparatus main body 21, and the dialing time corresponding to each telephone number is set at fixed time intervals.

On the other hand, when the dialing time comes, the corresponding telephone number is dialed by auto dialing. With this, if the other party goes off-hook, the memory 2
Transmission of the voice message stored in (voice synthesizer 11) is started (step 31). At this time, the presence or absence of the response voice of the other party is monitored, and if there is the response voice, the identification process for the speaker of the other party is performed (steps 32 and 33). In this case, the response voice signal Sr obtained from the 2-wire / 4-wire converter 12
Is after the echo component is canceled by the differentiator 13,
It is given to the first bandpass filter 3a and the second bandpass filter 3b, respectively. Then, the band pass filters 3a and 3b respectively filter the low frequency fa component and the high frequency fb component from the signal Sr to obtain the filtered signals Sa and Sb at the respective frequencies fa and fb. The obtained filtered signals (analog signals) Sa and Sb are A / D converter 5a,
After being converted into digital signals by 5b, they are given to the control unit 4, and the control unit 4 controls the magnitudes Pa of the filtering signals Sa and Sb at the respective frequency fa and fb components,
Find Pb. In this case, the control unit 4 samples the magnitudes of the filtered signals Sa and Sb a plurality of times (five times in the embodiment) over a certain period of time, and integrates the sampled values Ua ..., Ub ... As shown in FIG. Due to
The magnitudes Pa and Pb of the filtering signals Sa and Sb are obtained as integrated values (steps 51 and 52). When a predetermined time has passed and the speaker's response voice disappears, the ratio of Pa and Pb, that is, Pa / Pb is calculated, and this Pa / Pb is calculated.
Is temporarily stored as a speaker identification result (step 53,
54).

On the other hand, if there is a response voice of the other party during the transmission of the voice message, identification processing for the speaker of the other party is similarly performed (steps 34, 35, 36). In this case, first, a timer for measuring the set time Ts is started (step 61). Then, similarly to the identification processing in step 33, the magnitudes of the filtered signals Sa and Sb are sampled a plurality of times (five times) over a fixed time, and as shown in FIG. 4, the sampled values Ua ...
By adding Ub ..., the filtering signal S
The sizes Pac and Pbc of a and Sb are obtained as integrated values (steps 62 and 63). The set time Ts of the timer determines the number of times of sampling, and in the case of the embodiment, it is 5
Set to time up when sampling is completed. Therefore, when the timer times up, P
The ratio of ac to Pbc, that is, Pac / Pbc is calculated, and this Pac / Pbc is used as the speaker identification result (step 6).
4, 65).

The control unit 4 executes the identification process for the speaker at predetermined time intervals and monitors whether the identified speakers are different. Then, (Pa / Pb) <(Pa
c / Pbc), it is determined that the speakers are different, and the control unit 4 sends a voice message from the beginning. That is, the voice synthesizing unit 11 outputs a voice message from the beginning (steps 37, 31). This allows, for example, an adult to hear a voice message from the beginning, even if a child who cannot answer first picks up the handset of the telephone and turns into an adult who can answer in the middle.
Enables accurate and reliable telemarketing at all times.

When the other party does not respond even if the voice message is transmitted, when the voice message is completely transmitted (step 38) or step 3
When the voice message has been transmitted after the speaker is identified for the first time in step 3 (step 34), the tape recorders 25 are operated to start recording the response voice (step 39). At this time, the clocking of the set time Tr is started (step 40), and if there is a response voice of the other party, recording is performed, and if the set recording time has elapsed, the telemarketing operation is ended (steps 41, 45). ). On the other hand, when the set time Tr elapses without the other party's response voice, the timer times out (steps 41 and 42), and when the set recording time elapses, the telemarketing operation is terminated (step 44). , 45). Also, in the case where the recording time does not end even after the set time Tr has passed, and if there is a response voice of the other party during this time, the transmission of the voice message is started again from the beginning (steps 42, 43, 4).
5).

The embodiment has been described in detail above.
The present invention is not limited to such an embodiment.
For example, the frequency components can be any number of different frequency components. The frequency component is a concept including a frequency band component. On the other hand, as a speaker identification means, for example,
Such as identifying the absolute value of the magnitude of the filtered signal,
A more reliable identification may be made possible by combining other identification means. Furthermore, although the embodiments have illustrated bandpass filters, this does not preclude the use of lowpass or highpass filters. In addition, the detailed configuration, method, and the like can be arbitrarily changed without departing from the scope of the present invention.

[0023]

As described above, the telemarketing apparatus according to the present invention samples a plurality of filters for filtering different frequency components from the signal of the response voice and the magnitude of the filtered signal obtained from each filter, and Since the other party's speaker is identified at predetermined time intervals from the relationship of the magnitude of the filtering signal in the component, and a control unit is provided to send a voice message from the beginning if it is determined that the identified speaker is different.
Even if the speaker of the other party changes on the way, the voice message is transmitted from the beginning, and there is a remarkable effect that the telemarketing can be performed accurately and surely.

[Brief description of drawings]

FIG. 1 is a block circuit diagram of a main part of a telemarketing device according to the present invention,

FIG. 2 is an overall front view of the same telemarketing device,

FIG. 3 is a relationship diagram showing an integrated value of magnitudes of filtering signals for respective frequencies in the same telemarketing device,

FIG. 4 is a relationship diagram showing an integrated value of magnitudes of other filtering signals with respect to respective frequencies in the telemarketing device.

FIG. 5 is a flowchart showing the operation of the telemarketing device,

FIG. 6 is a flowchart specifically showing first-time speaker identification processing in FIG.

FIG. 7 is a flowchart specifically showing the n-th speaker identification process in FIG.

[Explanation of symbols]

1 Telemarketing device 2 Memory 3a Bandpass filter (filter) 3b Bandpass filter (filter) 4 Control part 5a Analog-digital converter 5b Analog-digital converter Sr Response voice signal Sa Filtering signal Sb Filtering signal fa Low frequency band fb High frequency Pa Pa magnitude of filtering signal (integrated value) Pac Magnitude of filtering signal (integrated value) Pb Magnitude of filtering signal (integrated value) Pbc Magnitude of filtering signal (integrated value) Ua ... Magnitude of filtering signal Sb is a sampled value Ub ... is a sampled value of the filtering signal

Claims (6)

[Claims] 1. A telemarketing device in which voice messages stored in a memory are sequentially sent to a party connected to a line by dialing, and a response voice of the other party is recorded. Multiple filters for filtering frequency components,
The size of the filtering signal obtained from each filter was sampled, and the speaker of the other party was identified at predetermined time intervals based on the relationship of the magnitude of the filtering signal in each frequency component, and it was determined that the identified speaker was different. If so, a telemarketing device comprising a control unit for sending a voice message from the beginning. 2. The telemarketing device according to claim 1, further comprising an analog-digital converter that converts a filtering signal obtained from each filter into a digital signal. 3. The telemarketing device according to claim 1, wherein the filter is a bandpass filter. 4. The telemarketing device according to claim 1, wherein the magnitude of the filtering signal is obtained by integrating values sampled a plurality of times. 5. The telemarketing device according to claim 1, wherein the different frequency components are a low frequency component and a high frequency component. 6. The telemarketing device according to claim 1, wherein the control unit identifies the speaker based on the ratio of the magnitude of the filtered signal in each frequency component.