JPH01317056A - Interactive automatic answering telephone set - Google Patents

Abstract

PURPOSE:To prevent a message recording enable time from lengthening by splitting the inside of a caller message recording part into two areas as a voice recording part during conversation and a message voice recording part and adding a message informing a fixed time length able to record the message to a reply message urging the message recording. CONSTITUTION:The inside of a caller message recording part 7 is split into two areas as an interactive voice recording part 7a during conversation recording the caller voice before the transmission of the reply message urging the message recording to the caller and the message voice recording part 7b recording the caller voice after the transmission of the reply message urging the message recording to the caller. Moreover, a message informing a fixed time length able to record the message is added to the reply message urging the message recording. Thus, the time length relating to the recordable time length of the message in the reply message sent finally in the automatic reply of an interactive type automatic answering telephone set and urging the caller for the message recording is always the same time length. Thus, the message recording enable time length is prevented from being lengthened.

Description

[Detailed Description of the Invention] [Industrial Field of Application] The present invention provides an interactive answering machine that records the caller's voice while exchanging messages between the caller and the telephone when automatically receiving a call. It is related to.

[Prior Art] Conventionally, a plurality of pre-recorded response messages are sent out one after another in response to the caller's voice, thereby making the caller aware that the answering machine is automatically answering the caller. An interactive answering machine is known that records messages without having to use the telephone. This interactive answering machine is different from previous answering machines that automatically answer messages such as ``the phone owner is not available'' or ``reminding you to record the message,'' but only a single response message is sent from the answering machine. In contrast, there is a high probability that the caller will make some kind of utterance. Taking advantage of this feature, there is an interactive answering machine developed for the purpose of countering nuisance calls. The response message of this interactive answering machine consists of two parts.
``Yes, this is ○O''), and the second response message is a message stating ``the telephone owner is absent'' or ``requesting recording of the business''. At the time of automatic response, the first response message and the second response message are sent to the line with a certain pause period in between. The telephone owner can confirm the caller from the caller's voice uttered after sending the first response message and determine whether the incoming call is a nuisance call or not. However, in this method, the timing of sending the second response message is
Since it is not synchronized with the end of the caller's message, the caller will be aware of the unnaturalness of the conversation when the caller finishes speaking or when the answering machine starts sending out the second response message. It is difficult to feel this and carry out the subsequent dialogue naturally. Therefore, by having a means to detect the voice from the caller, it can detect when the caller has finished speaking, and send the next response message in time with the timing when the caller has finished speaking, making the dialogue more natural. There is an interactive answering machine that can be used for a long time (Japanese Patent Application Laid-Open No. 189057/1983). Furthermore,
The content of the caller's voice is estimated based on the length of the caller's voice and whether or not the caller responds to the caller's response, and if the caller does not say his or her name, the caller's name is used. Interactive answering machines have also appeared that have the ability to at least hear the name of the caller while allowing natural dialogue to occur through actions such as sending out messages asking for the caller's name.
No. 2052).

FIG. 4 is a diagram showing the operational flow of a conventional interactive answering machine that performs a dialog response while estimating the utterance content of the caller's voice, and FIGS. 5(a) and 5(b).

(c) is a diagram showing an example of a dialog response performed between the answering machine and the caller. In this conventional example, when a ringing signal from the central office line is detected, the line is closed, and the call is automatically received, a first response message such as "Yes, this is OO Shoji" is sent out, and the caller's message is Start recording. During this time, the caller's voice is detected, the caller's message is judged to have finished speaking, and the above recording is interrupted, and the time from the time the caller speaks to the end is monitored to shorten the length of the caller's message.
.

It is classified into three levels: "medium" and "long", and responds as follows according to each case. If it is "short", it is presumed that there was a message from the caller such as "Moshimoshi" as shown in Figure 5(b), a second response message "Yes" is sent, and the above message is sent again. The process from recording the caller's message to determining the length of the message is repeated. 0 if the caller's message length is "medium" or 5 if the caller's message is "My name is Sato" in Figure 5(b) or the second caller's message. The caller presumes that the message is "Mr. Tanaka is here?" as shown in FIG. Here, in the case of FIG. 5(c), it is presumed that there is no subsequent message from the caller (silence), so
The presence or absence of a response from the caller is detected by time monitoring, etc. When there is a response, the system infers Fig. 6 (b), detects the end of the caller's utterance, interrupts the above recording, and sends the fourth response message ``I'm not home at the moment.'' Please record your message.'' After recording the caller's message (message) until the end of the utterance, the line is released and the call ends. If there is no response, the caller assumes the situation shown in FIG. 5(c), interrupts the recording, and sends out a third response message, ``How are you?'', and is equipped with the ability to hear at least the caller's name. Subsequently, the recording of the caller's message is started and interrupted when the utterance is finished, and then, similarly to the above, the process starts from sending out the fourth message shown below to releasing the 9th line for recording the business, and ends. In addition, if the message length mentioned above is "long", it is assumed that the caller's message in Figure 5 (a) is "I'm Sato, but Mr. Tanaka, I'd like to help you", and execute the steps below and finish. do.

As mentioned above, the caller's message during a conversation contains important information about the caller, such as the caller's name and the name of the person with whom the caller has business matters. . Therefore,
As mentioned above, in these conventional interactive answering machines, the caller's message during this period is also recorded and stored in the caller's message recording section one by one along with the message. However, in any case, there is no system that prompts the caller to record a message and then responds in an interactive manner. This is because callers generally have a wide range of business matters, so it is extremely difficult to estimate the content of a caller's message based on information such as voice length. This is because even if techniques such as natural language processing are used, it is difficult to understand the meaning of the voice in which the caller describes the business.

[Problems to be Solved by the Invention] However, in the above-mentioned conventional interactive answering machine, the caller's voice is recorded in the caller's message recording section during the conversation before the message is sent to remind the user to record the message. Because the message is recorded, the length of time that the message can be recorded varies depending on the length of the caller's message recorded during the previous conversation. There was a problem in that it was difficult to mention the possible length of time.

FIGS. 6(a), (b), and (C) are explanatory diagrams showing changes in the possible recording time of business matters in the conventional example, and FIGS. 5(a), (b), and (c) are respectively This figure shows the length of time that the caller's message recording unit can record the message voice when a conversation occurs between the caller and the answering machine. As mentioned above, in the interactive answering machine, Figs. 5(a) and (b)
, (c), in order to automatically respond by estimating the conversation with the caller, the recording time length of the caller's message (recording time during the conversation) until the fourth response message is sent to prompt the caller. Long) TCNIJ is shown in Figures 6(a) and (b).

As shown in (c), TcNu(a)<']'0+u(b
)<Tcwu(c). On the other hand, since there is a limit of 8 times on the recording medium that records all caller messages including the recording time length, the remaining message length after subtracting the recording time during conversation from the total recording time The recording time length TAFTCNLI will also vary, as shown in Figure 6(a).
, (b) and (c), Thvrc,lu (a
) >TAvTcsu (b) >TAvtcsu
(C). For this reason, during the fourth response message to remind recording of a message, the recordable time TAF? (:N
To mention O, detect the length of the unused area remaining in the caller voice recording section, calculate the recordable time length of the message voice according to it, and then adjust the recording time of the message according to the result. It is necessary to change the content regarding the recording time length of the business voice in the reminder message. Possible ways to do this include changing the content of the voice regarding the available time for recording messages that is sent during the response message urging recording of messages, depending on the amount of unused area in the caller's message recording section. , it is necessary to newly provide a voice storage device or the like for producing several types of temporal sounds, which is not practical.

Since a general answering machine first prompts the caller to record the message, the length of time that the message can be recorded is fixed, and the message is recorded as a guideline for the caller to summarize and speak the message. The length of time that voice can be recorded can be easily communicated to the caller. On the other hand, in an interactive answering machine that responds to the caller while determining whether the caller has finished speaking, a response message (corresponding to the fourth response message in Figure 5) that prompts the caller to record the message is sent. In particular, it is necessary to mention the length of time that the business audio can be recorded. This is because in many cases, the caller does not realize that what he or she has said up to that point is being recorded, so if you record the message without a time limit, the caller may have to start speaking again by mentioning the caller's name, etc. There is a high probability that the recording of the message will not end even after the hard time limit has been exceeded, and in such a case, even if you play back the recorded message, the last part of the message will be This is because there is a high possibility that the telephone owner will not be able to fully understand the reproduced message from the caller because it is not recorded.

The present invention was devised to solve the above-mentioned problems, and is an interactive answering machine that allows the length of time for which a message can be recorded to be easily mentioned in a response message that urges recording of a message while giving a dialog response. The purpose is to provide telephones.

[Means for Solving the Problems] The configuration of the interactive answering machine of the present invention to achieve the above object is to automatically receive a call and send a plurality of response messages stored in a response message storage section from the caller. In an interactive answering machine that sends out different messages depending on the message, records the caller's voice in the caller message recording unit, and automatically responds in an interactive manner, the inside of the caller message recording unit A dialogue voice recording unit that records the caller's voice before sending a response message to the caller requesting that the caller record the message; It is characterized in that it is divided into two areas: a business voice recording section for recording the caller's voice, and a message conveying a fixed length of time during which the business can be recorded is added to the response message urging recording of the business.

However, according to the means of the present invention, an area for a certain period of time is always prepared for recording the caller's business.

Therefore, in the response message that is sent at the end of an automatic answering machine of an interactive answering machine and that reminds the caller to record the message, the reference to the length of time that the voice message can be recorded may always be the same length. It turns out. As a result, when a telephone owner pre-records a response message, he or she can always state the same length of time for recording the message, making it easier to mention it and reducing the burden on the telephone owner. Reduced. At the same time, during automatic answering, the caller knows the available time for recording the message, which can be used as a guideline for determining the speaking speed and content when recording the voice of the message. This reduces the possibility of running out of recordable time.

[Example] Hereinafter, an example of the present invention will be described in detail based on the drawings.

FIG. 1 is a block diagram of an interactive answering machine showing one embodiment of the present invention. 1 is an incoming call detection unit that is connected to the central office line, L, and detects a calling signal from the central office line, and 2 is a microcomputer that makes judgments based on signals from each unit and issues control commands and instructions to each unit. A control unit that enables automatic call reception and interactive automatic response. The incoming call signal detected by the incoming call detection section l is sent to the control section 2. Reference numeral 3 denotes a loop opening/closing section which opens/closes the telephone line and the DC loop according to instructions from the control section 2, and 4 denotes a speech circuit section connected to the central office line L1Lt via the loop opening/closing section 3. The telephone call circuit section 4 includes receiving terminals R1 and R2 and transmitting terminals T + and T.
It has the function of separating a bidirectional telephone line into a unidirectional receiving line and a sending line.

Reference numeral 5 indicates a response message sending section which is connected to the transmitting terminals T, , T, of the telephone communication circuit section 4 and sends a response message to the central office line, . This is a response message storage section that stores response messages. The response message sending section 5 has a function of sending out a response message from the response message storage section 6 instructed by the control section 2 at a timing instructed by the control section 2. 7 is connected to the receiving terminals R,, rl, of the telephone communication circuit section 4, and records the caller's message sent via the central office line, , t, at the timing instructed by the control section 2. This is a caller message recording section that is started or interrupted, and its internal recording medium is divided into two areas: a conversation voice recording section 7a and a business voice recording section 7b. Reference numeral B indicates a voice detection unit which is also connected to the receiving terminal R1゜RO of the telephone call circuit unit 4 and detects the utterance/end of the caller's voice and sends it to the control unit 2; 9 indicates the continuation of the silent state of the caller's voice; /Clock unit 10 for measuring and outputting the continuation of the sound state according to instructions from the control unit 2
is a recordable time length monitoring unit that monitors the recordable time length of the caller message recording unit 7 according to instructions from the control unit 2.

FIG. 2 is a diagram showing an example of the internal configuration of the caller message recording section. Various recording media can be used, such as a tape recorder's recording tape or a semiconductor recording/reproducing device.This is divided into two areas, and one area is used to remind the caller to record the message. The dialogue voice recording section 7a records the caller's voice before sending the response message, and the other area records the caller's voice after sending the response message urging the caller to record the business. This is referred to as a business voice recording section 7b. Here, the recordable time length of the dialog voice recording section 7a among the total recordable time length T ALL: TcNu
is set to the maximum length of voice expected to be uttered by the caller during a normal dialog response. In this way, the recordable time length TArycNu is set to a fixed length. Table 1 is an example of a response message that is stored in advance in the response message storage unit 6, and the fourth response message that urges the caller to record a message conveys the fixed-length message recordable time length TAFTcIIIJ. Add the message "Recording time is (T□.)." However, TMES
S≦TAFTCNtl. (Left below) Table 1 Example of response message The operation of the embodiment configured as described above will be described.

FIG. 3 is a flowchart for explaining the operation of this embodiment, and the operation will be explained with reference to FIG. First, when there is an incoming call, the incoming call detection section 1 detects this and sends an incoming signal to the control section 2. When the control section 2 receives this incoming call signal, after a predetermined period of time has elapsed, the control section 2 operates the loop opening/closing section 3 to close the loop (connecting the line), and ends the automatic incoming call operation. When the automatic call receiving operation is completed, the control section 2 selects the first response message "Yes, this is OO Trading" from the response message storage section 6, and
The response message output section 5 sends the message to the central office line ・L via the telephone communication circuit section 4. When the transmission of the first response message is completed, the caller message recording section 7 is activated.
The caller's voice appearing on the central office lines L1 and L is transmitted to the telephone communication circuit section 4.
The message is recorded in the dialog voice recording section 7a of the caller's message recording section 7 via the caller message recording section 7. In this state, the recordable time length monitoring section 10 continues to monitor the recordable time length of the dialog voice recording section 7a. Then, when the available recording time of the dialog voice recording unit 7a runs out (■), the recording operation of the caller message recording unit 7 is interrupted, and the control unit 2 adds the fourth response message to the next response message to be sent. select. At the same time, in this state, the voice detection section 8 measures the presence or absence of the caller's voice, and the clock section 9 measures the duration of the sound state/silence state. When the duration of silence exceeds a certain threshold (TED), the caller's utterance is not considered to have ended, and the recording operation of the caller's message recording unit 7 is interrupted. , the control unit 2 immediately prepares to send the next response message. If the response message sent immediately before is the first response message, the next response message and The subsequent operation flow branches into the following three ways.

(1) If the caller's message is long, it is assumed that the caller has already uttered the caller's name or the name of the person with whom the caller has a message, so the first response message is selected as the next response message.

(2) If the caller's message is short, it is assumed that the caller has only uttered something like "Moshi-moshi", so as a response, a second response message "
After sending "Yes", the same operation as after sending the first response message is performed again.

(3) If the caller's message is of medium length, it is considered that the caller only announced the name of the caller or the name of the other party who had business to do. As a response, the second response message “
"Yes" is sent, and the recording of the caller's voice is resumed in the dialog voice recording section 7a of the caller's message recording section.

In this state, if the caller's voice becomes active, the first
The message sent by the caller immediately after the response message is considered to have only announced the name of the caller, so after confirming that the caller has finished speaking according to the method for confirming that the caller has finished speaking, the caller's message is sent immediately after the response message. The recording operation of the recording section 7 is interrupted, and the fourth response message is selected as the next response message. In addition, when the recording time length monitoring unit 10 detects that the recording time of the dialog voice recording unit 7a has run out while recording the caller's voice, it immediately stops the recording operation of the caller message recording unit 7. The process is interrupted, and the fourth response message is selected as the next response message to be sent. On the other hand, in this state, if the caller's voice does not become active and the silence continues for more than a certain threshold (TNA), the caller's message immediately after the first response message is the name of the person with whom the caller has a message. It is thought that the caller's name has not yet been announced, as it is only a notification. Therefore,
■As shown below, after the recording operation of the caller message recording unit 7 is interrupted, the third response message “Excuse me, but which one do you have?” is sent, and the caller message recording unit records the voice during the conversation. The recording of the caller's voice is resumed in section 7a. After confirming that the caller has finished speaking according to the caller's voice completion confirmation method described above, the recording operation of the caller's message recording section is interrupted, and the fourth response message is selected as the next response message. In addition, when the recording time length monitoring section 10 detects that the recording time of the dialog voice recording section 7a has run out while recording the caller's voice, it immediately stops the recording operation of the caller's message recording section 7. The process is interrupted, and the fourth response message is selected as the next response message to be sent.

After the above processing, ■The next response message shown below is the fourth response message: ``I'm away at the moment.I'm going to record my business, so please talk to me.The recording time is (TME8g). '' is selected, after sending this message, the caller message recording section 7 is activated and recording of the caller's voice is resumed. The caller's voice at this time is recorded in the message voice recording section 7b of the caller's message recording section 7, since it is immediately after the caller is prompted to record the message. At this point, the caller's voice until caller voice recording starts is as follows:
All caller message recording unit 7 dialogue voice recording unit 7a
Therefore, nothing is recorded in the message voice recording section 7b of the caller's message recording section. That is,
The recordable time length of the business voice is equal to the recordable time TAFTCNU of the business voice recording section of the caller's message recording section. Therefore, the relationship between the recordable time Tsess of the business voice mentioned in the fourth response message and the recordable time TAFTCNU of the business voice recording section of the caller's message recording section is fixed as T,4m5s=TAFycwu. It can be made long. Of course, take your time and TM! lss
< T AvrcNu may also be used. In this state, the recordable time length monitoring unit 10 monitors the recordable time of the business voice recording unit 7b. Then, when it is detected that the recordable time of the business voice recording section 7b has run out or that the caller's voice has ended, the loop opening/closing section 3 opens the line and ends the automatic response operation.

As described above, in this embodiment, the caller message recording section 7 is divided into two areas: the dialog voice recording section 7a and the business voice recording section 7b, and the recording time during the conversation is monitored, and the message recording section 7 is divided into two areas: the conversation voice recording section 7a and the business voice recording section 7b. By interrupting the recording when the time comes, a certain length of time for recording the business voice is secured. As a result, this embodiment allows the caller to mention the available recording time as a guideline in the response message urging the caller to record the business.

In addition, when the recording medium of the caller message recording unit 7 in the above embodiment is divided into two areas, one recording medium may be divided into two usage areas, or depending on the recording medium, separate recording media may be used. This can also be achieved by dividing it into two parts. As described above, the present invention can be applied in various ways and can take various embodiments in accordance with its gist.

[Effects of the Invention] As is clear from the above description, according to the interactive answering machine of the present invention, an area of a certain length of time is always prepared for recording the voice of the caller. Therefore, the utterance regarding the length of time that the voice message can be recorded in the response message that is sent at the end of the automatic answering machine of an interactive answering machine to remind the caller to record the message may always be the same length. It turns out. As a result, when the telephone owner records a response message in advance, he/she only needs to always state the same recordable time for the message, thereby reducing the burden on the telephone owner. At the same time, during automatic answering, the caller knows the available time for recording the message and can use this as a guideline for determining the speaking speed and content when recording the message. This reduces the possibility of running out of available recording time.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of an interactive answering machine showing one embodiment of the present invention, FIG. 2 is a diagram showing an example of the internal configuration of a caller message recording section, and FIG. 3 is a diagram for explaining the operation of this embodiment. Flowchart, Fig. 4 is a diagram showing the operation flow of a conventional interactive answering machine, and Figs. 5 (a), (b), and (c) show the conventional interactive answering machine and the caller's dialog response. FIGS. 6(a), 6(b), and 6(c), which show an example, are explanatory diagrams showing changes in the length of time during which business voice recording is possible in a conventional example. l... Incoming call detection unit, 2... Control unit, 3... Loop opening/closing unit, 5... Response message sending unit, 6... Response message storage unit, 7... Caller message recording Department,
7a... Dialogue voice recording unit, 7b... Business voice recording unit, 8... Voice detection unit, 10... Recordable time length monitoring unit.

Claims (1)

[Claims] (1) Automatically receives a call, sends out multiple response messages stored in the response message storage section depending on the message from the caller, and records the caller's voice in the caller message recording section. However, in an interactive answering machine that automatically responds in a conversational format, the inside of the caller message recording section records the caller's voice before sending a response message urging the caller to record the message. A response prompting the caller to record the business is divided into two areas: a voice recording section and a business voice recording section that records the caller's voice after sending a response message urging the caller to record the business. An interactive answering machine characterized in that a message indicating a fixed length of time during which a message can be recorded is added to the message.