Auto-call processing method, computer program, and auto-call processing system

Abstract

To provide a new technology for suitably implementing an auto-call using an automated voice.SOLUTION: In an auto-call processing method for playing back a message via auto-call, one or more computers make an auto-call to a target person, play back a message to the connected target person, and also obtain the target person's voice to determine the connection status, and play back a new message from the beginning when it is determined that the connection status is an answering machine.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to an auto call processing method, a computer program, and an auto call processing system. [Background technology]

[0002] 2. Description of the Related Art A number of auto-call processing techniques have been proposed for implementing auto-calls (automatic outgoing calls) by automated voice, with as little need for intervention by a telephone operator as possible.

[0003] An example of such an auto-call processing technology is proposed, for example, in Patent Document 1. This proposed technology has been realized by the applicant as a survey service for conducting surveys (questionnaires) in the form of questions and answers. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2019-079411 A

[0005] If the connection with the target person is via an answering machine, the target person will hear a mechanical beep before the answering machine starts recording. Therefore, while the automated voice message is being played back, the answering machine may start recording from an unintended part of the message. Summary of the Invention [Problem to be solved by the invention]

[0006] An object of the present invention is to provide a novel technique for suitably implementing an auto-call using an automated voice. [Means for solving the problem]

[0007] In order to solve the above problems, the present invention provides an auto-call processing method for playing back a message via auto-call, in which one or more computers make an auto-call to a target person, play a message to the connected target person, and also obtain the target person's voice to determine the connection status, and if it is determined that the connection status is an answering machine, play a new message from the beginning.

[0008] With this configuration, even if the phone is connected to an answering machine, an autocall can be implemented that allows the voice information of a message to be transmitted from an intended location.

[0009] In a preferred embodiment of the present invention, when the computer determines that the call is an answering machine, the message that has been being played back up to that point is replayed as a new message to be played back from the beginning. With this configuration, it is possible to play back messages in accordance with the purpose of the auto-call service.

[0010] In a preferred embodiment of the present invention, when the computer determines that the call is an answering machine, a new message is played back from the beginning, which is different from the message that was being played back up until that point. With this configuration, it is possible to play back messages in accordance with the purpose of the auto-call service.

[0011] In a preferred embodiment of the present invention, the computer stores standard message data for playing back standard messages and answering machine message data for playing back messages for answering machines, and when the computer makes an auto-call to a target person and is connected to the target person, it plays back the standard message to the connected target person, and if it determines that the call is an answering machine, it stops playing back the standard message and plays back the answering machine message from the beginning again. With this configuration, it is possible to play back messages in accordance with the purpose of the auto-call service.

[0012] In a preferred embodiment of the present invention, an arbitrary waiting time is provided between when the computer determines that the connection status is an answering machine and when the computer starts reproducing a new message from the beginning. With this configuration, it is possible to newly play back a message from the beginning.

[0013] In a preferred embodiment of the present invention, the computer detects an answering machine connection using a detection model that has been trained on one or more types of mechanical sound features, and the mechanical sound features obtained by performing frequency analysis on the input subject's voice. With this configuration, connection to an answering machine can be detected in an appropriate manner.

[0014] In a preferred embodiment of the invention, the message is recorded voice data. With this configuration, it is possible to play back messages in accordance with the purpose of the auto-call service. Effect of the Invention

[0015] The present invention can provide a novel technique for suitably implementing an auto-call using an automated voice. [Brief description of the drawings]

[0016] [Figure 1] FIG. 1 is a block diagram showing a configuration of an auto-call processing system according to an embodiment. [Diagram 2] FIG. 2 is a block diagram showing the hardware configuration of an administrator terminal device and a subject terminal device according to an embodiment. [Diagram 3] FIG. 2 is a block diagram showing a hardware configuration of an autocall processing device according to an embodiment. [Figure 4] FIG. 2 is a block diagram showing the functional components of an autocall processing device according to one embodiment. [Diagram 5] FIG. 4 is a diagram for explaining a target person list according to an embodiment. [Figure 6]FIG. 4 is a diagram for explaining an auto-call process according to an embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0017] The present invention will be described in more detail below with reference to the accompanying drawings. Although the drawings show a preferred embodiment, the present invention can be implemented in different forms and is not limited to the embodiment described in this specification. In this embodiment, the configuration and operation of an auto-call processing system will be described, but the same effects can be achieved by a method executed by a computer constituting the auto-call processing system, a computer program executable by a computer, etc. The computer program may be provided as a non-transient recording medium readable by a computer.

[0018] <Auto call processing system> Referring to Figure 1 showing the system configuration in one embodiment, the auto call processing system 1 includes an auto call processing device 2, a terminal device used by an administrator (administrator terminal device) 3, a terminal device used by a target person (target person terminal device) 4, and a communication network 5.

[0019] The communication network 5 is composed of a telephone network including a switchboard, and an IP (Internet Protocol) network such as the Internet including a router, and accommodates an auto-call processing device 2, an administrator terminal device 3, and a target user terminal device 4. This communication network 5 accommodates a large number of administrator terminal devices 3 and target user terminal devices 4, but only representative administrator terminal devices 3 and target user terminal devices 4 are shown here.

[0020] The communication network 5 also includes a VoIP (Voice over Internet Protocol) gateway that connects the telephone network and the IP network and converts between voice signals and IP packets. In the following explanation, the inclusion of the communication network 5 will be omitted unless it is unclear.

[0021] The subject user terminal device 4 may have a single or composite configuration including at least one of a fixed telephone terminal TEL, a mobile phone terminal MT (including a smartphone SP), and a computer terminal CP. The manager user terminal device 3 includes a computer terminal CP. The subject user terminal device 4 (4A) illustrated here is a single configuration including a mobile phone terminal MT, the subject user terminal device 4 (4B) is a single configuration including a computer terminal CP, and the subject user terminal device 4 (4N) is a single configuration including a fixed telephone terminal TEL. However, the configuration of the subject user terminal device 4 is not limited to the illustrated configuration as long as it has a voice communication function and a data communication function.

[0022] The target user terminal device 4 is assigned a telephone number and / or an IP address for voice communication and / or data communication with the autocall processing device 2.

[0023] The computer terminal CP constituting the administrator terminal device 3 and the subject terminal device 4 is, for example, a personal computer. The computer terminal CP constituting the subject terminal device 4 (4B) is equipped with VoIP software to use IP telephones. This VoIP software converts between IP packets and voice signals that the computer terminal CP transmits and receives via a broadband line, enabling IP telephone calls using speakers, microphones, etc.

[0024] In further detail, the manager terminal device 3 (3A, 3B, ... 3N) and the target user terminal device 4 (4A, 4B, ... 4N) in the above-mentioned auto call processing system 1 include hardware components as illustrated in Figure 2.

[0025] In other words, the administrator terminal device 3 and the target user terminal device 4 each have, as hardware components, a CPU (Central Processing Unit) 11 as a processor, a RAM (Random Access Memory) 12 as working memory, and a ROM (Read Only Memory) 13 that stores a boot program for startup.

[0026] The administrator terminal device 3 and the target user terminal device 4 further include a non-volatile flash memory 14 that rewritably stores an OS (Operating System), application programs, and various information (including data), a communication control unit 15, and a communication interface (IF) unit 16 such as a NIC (Network Interface Card).

[0027] In addition, the manager terminal device 3 and the subject user terminal device 4 further include a display control unit 17, a display unit 18, and an information input / designation unit 19. The "display" in the subject user terminal device 4 is an audible display through a speaker.

[0028] The auto-call processing device 2 is an auto-call service providing server with an automated voice, operated and managed by an ISP (Internet Service Provider) or the like, and performs the auto-call processing described in detail below. The automated voice is played back using recorded voice data and / or synthetic voice data generated from a script. For example, an IPS or the like uses the auto-call processing device 2 to provide a policy issuing auto-call service, a product sales auto-call service, and the like.

[0029] This auto call processing device 2 has both a data communication function and a voice communication function, and includes hardware and functional components as shown in FIG.

[0030] That is, the auto-call processing device 2 includes, as hardware components, a CPU 21 as a processor, a RAM 22 as a working memory, and a ROM 23 that stores a boot program for start-up.

[0031] The autocall processing device 2 further includes a nonvolatile flash memory 24 that rewritably stores an OS, application programs, and various information (including data), a communication control unit 25, and a communication interface (IF) unit 26 such as a NIC.

[0032] As shown in FIG. 4, the auto-call processing device 2 includes, as functional components described in detail later, a registration processing unit 201, an outgoing call processing unit 202, an analysis processing unit 203, VoIP software 204, and databases DB1, DB2, and DB3.

[0033] Here, VoIP software 204 for utilizing IP telephones is stored in the flash memory 24. Also, databases DB1, DB2, and DB3 are configured in the flash memory 24.

[0034] To logically realize the above-mentioned other functional components (201-204) in the auto-call processing device 2, an auto-call processing program is installed as an application program in the flash memory 24. Then, in the auto-call processing device 2, when the power is turned on, the processor (CPU) 21 constantly loads this processing program in the RAM 22 and executes it.

[0035] <Auto call processing device details> Next, with reference to FIG. 1 and related figures (FIGS. 5 and 6), the auto-call processing device 2 that performs an auto-call (automatic outgoing call) by an automated voice without the need for the intervention of a telephone operator will be described in more detail.

[0036] The registration processing unit 201 registers auto-call information for performing auto-call processing. In this embodiment, the registration processing unit 201 registers the message data, the target person list, and the setting data received from the manager terminal device 3 in a database as auto-call information.

[0037] The message data is voice information for outputting (playing) a message when an auto-call is connected, and the registration processing unit 201 registers the standard message data, or the standard message data and the message data for an answering machine in the database (message data) DB1. The standard message data indicates a message that is played when an auto-call is made and a connection is made with the target party, and the message for an answering machine indicates a message that is played when an answering machine determination is made, which will be described later. The standard message may be used as the message for an answering machine that is played when an answering machine is connected. That is, the registered standard message data may be used as the message data for an answering machine. In this embodiment, the message data is a recorded voice file, but in order to automatically play a synthetic voice for the target party, a script file written in a predetermined markup language (for example, VXML (Voice Extensible Markup Language)) may be stored in the database DB1 as message data.

[0038] The target person list is a collection of telephone numbers related to the target person's terminal device 4 to which calls are to be made, and the registration processing unit 201 registers the target person list in a database (target person list) DB2. As illustrated in FIG. 5, an arbitrary number of target telephone numbers, such as target telephone numbers (e.g., mobile phone number "090-XXXX-0001", IP telephone number "050-YYYY-0001", landline telephone number "03-ZZZZ-0001") corresponding to the target person's terminal device 4 (4A, 4B, . . . 4N), are registered in the database DB2. In this embodiment, the registration processing unit 201 is described as registering a target person list that has been created in advance in the database DB2 via the administrator's terminal device 3, but the registration processing unit 201 may generate a target person list based on information acquired from a telephone directory or the like, and register the target person list in the database DB2.

[0039] The setting data is data indicating the settings of the auto-call, and the registration processing unit 201 registers the schedule such as the scheduled start date and time, the identification information of the message data used for the auto-call, and the identification information of the target person list in the database (setting data) DB3. The identification information of the message data is, for example, a path indicating the storage location of the message data or an ID assigned to the message data. The identification information of the target person list is, for example, a path indicating the storage location of the target person list or an ID assigned to the target person list.

[0040] The outgoing call processing unit 202 controls the transmission of voice information for voice output to the target person's terminal device 4A to the target person's terminal device 4N based on the message data. That is, the outgoing call processing unit 202 controls the target person's terminal device 4n (n=A, B, C,...N) to play back the message by referring to the databases DB1 to DB3.

[0041] With reference to the setting data registered in database (setting data) DB3, an auto-call by automated voice is started at the scheduled start date and time (see process S601 in FIG. 6). As the auto-call by automated voice, first, the outgoing call processing unit 202 acquires the outgoing call target telephone number (here, mobile phone number "090-XXXX-0001") corresponding to the target person's terminal device 4A (here, mobile phone terminal MT) from the target person list corresponding to the setting data registered in database (target person list) DB2 (see process S602 in FIG. 6), and transmits (originates a call) to the target person's terminal device 4A (see process S603 in FIG. 6).

[0042] If the target user's terminal device 4A does not respond to this call for a certain period of time (see process S604 (NO) in FIG. 6), the call processing unit 202 disconnects the call assuming that the target user's terminal device 4A did not answer (see process S609 in FIG. 6), and then makes a call to the next target telephone number (see process S610 (NO) in FIG. 6).

[0043] When the target person's terminal device 4A responds to this call (see process S604 (YES) in FIG. 6), the outgoing call processing unit 202 plays a standard message corresponding to the setting data and the call target list registered in the database (message data) DB1. That is, the outgoing call processing unit 202 transmits voice information based on the standard message data to the target person's terminal device 4A as an audible frequency band signal (see process S605 in FIG. 6). In the case of the policy outgoing auto-call service, the outgoing call processing unit 202 plays a standard message such as "Hello, I'm Ascend Taro, running for the XX election. In this election..."

[0044] The analysis processing unit 203 acquires the voice of the subject and judges the connection state. That is, the analysis processing unit 203 receives voice information (voice of the subject) which is an audible frequency band signal transmitted from the subject's terminal device 4 and detects mechanical sounds (see process S606 in FIG. 6). The voice of the subject is, for example, voice information such as "We are not on call right now. If you have any questions, please leave a message after the beep. Beep (mechanical sound)...", and when the analysis processing unit 203 detects a mechanical sound from the voice of the subject, it judges that the connection state is an answering machine.

[0045] The analysis processing unit 203 detects an answering machine connection using a detection model that has learned one or more types of mechanical sound features and mechanical sound features obtained by frequency analysis of the input target person's voice. The mechanical sound features are, for example, fundamental frequencies, frequency spectra, Mel-frequency cepstrum coefficients, logarithmic Mel filter bank features, etc. obtained by frequency analysis of voice information containing mechanical sounds, or parameters obtained by using at least one of these. The detection model is, for example, an acoustic model such as a hidden Markov model, a pattern matching dictionary model, or a deep learning model such as CNN or RNN.

[0046] When the connection state is determined to be an answering machine, the outgoing call processing unit 202 plays a new message. Here, the outgoing call processing unit 202 plays a different message from the message that has been played up until then as the new message to be played. When the analysis processing unit 203 determines that the connection state with the target person's terminal device 4A is an answering machine (see process S606 (YES) in FIG. 6), the outgoing call processing unit 202 plays an answering machine message corresponding to the setting data and the outgoing call target list registered in the database (message data) DB1. That is, the outgoing call processing unit 202 transmits voice information based on the answering machine message data as an audible frequency band signal to the target person's terminal device 4A (see process S607 in FIG. 6). In the case of the policy outgoing auto-call service, the outgoing call processing unit 202 plays an answering machine message such as, for example, "Hello, sorry to bother you during your busy schedule. I'm Ascend Taro, running for the XX election. In this election..."

[0047] Here, the outgoing call processing unit 202 provides an arbitrary waiting time from when the analysis processing unit 203 determines that the connection status with the subject's terminal device 4A is an answering machine until a new message is played back. The waiting time is a predetermined time or more, for example, one second or more, so that the new message played back after the analysis processing unit 203 determines that the message is an answering machine (detects mechanical sounds) is recorded as an answering machine in the subject's terminal device 4A from the beginning. The waiting time is also within a predetermined time, for example, four seconds, so that the answering machine recording does not end automatically.

[0048] When the call processing unit 202 loses connection with the target person's terminal device 4A (see process S608 in FIG. 6), it repeatedly performs the same call processing by multiplexing using at least one communication line and at least one channel until it has transmitted voice information to all of the target person's terminal devices 4A-N (see process S610 in FIG. 6).

[0049] <Effects of one embodiment> The auto-call processing device 2 in the auto-call processing system 1 according to the embodiment described above performs an auto-call by automated voice without the need for the intervention of a telephone operator. This auto-call processing device 2 detects whether the connection state of the target user terminal device 4 is an answering machine, and when it detects that it is an answering machine, it plays back a new message. This auto-call processing device 2 also plays back a new message for the target user terminal device 4 connected to the answering machine.

[0050] Therefore, in this auto-call processing device 2, even when connected to an answering machine, it is possible to implement an auto-call that enables the transmission of voice information of a message from an intended position (the beginning of the message).

[0051] <Modification> In the auto-call processing device 2 in the embodiment described above, when the analysis processing unit 203 determines that the connection state with the subject's terminal device 4 is an answering machine, the outgoing call processing unit 202 transmits voice information based on the message data for the answering machine as an audible frequency band signal to the subject's terminal device 4. However, when the outgoing call processing unit 202 determines that the connection state is an answering machine, it may replay (replay) the message that was being played back as a new message from the beginning. That is, the outgoing call processing unit 202 may transmit voice information based on the standard message data to the subject's terminal device 4 from the beginning again. In this case, the above-mentioned waiting time may be provided. [Explanation of symbols]

[0052] 1 Auto call processing system 2 Auto call processing device 3. Administrator terminal device 4. Target User Terminal Device 4A User terminal device 4B User terminal device 4N Target User Terminal Device 5. Communication Network DB1 database (message data) DB2 Database (Target List) DB3 database (configuration data) TEL Fixed line telephone terminal MT mobile phone terminal SP Smartphone CP Computer Terminal

Claims

1. A method for processing an auto-call to play back a message by an auto-call, comprising the steps of: One or more computers make an auto-call to the target person, play a message to the connected target person, and obtain the target person's voice to determine the connection status; When it is determined that the connection state is an answering machine, the auto-call processing method plays back the message from the beginning.

2. 2. The method of claim 1, wherein when said computer determines that the call is an answering machine, the message that has been being played back up to that point is played back again as the new message to be played back from the beginning.

3. 2. The auto-call processing method according to claim 1, wherein, when said computer determines that the call is an answering machine, a new message is played back from the beginning, which is different from the message that has been played back up until that point.

4. the computer stores standard message data for playing standard messages and answering machine message data for playing answering machine messages; The computer performs an auto-call to the target person, and when a connection is made with the target person, the computer plays the standard message to the connected target person; 2. The auto-call processing method according to claim 1, further comprising the steps of: if it is determined that the call is an answering machine, stopping the reproduction of the standard message and reproducing the message for the answering machine from the beginning.

5. 2. The auto-call processing method according to claim 1, wherein said computer provides an arbitrary waiting time from when it determines that said connection state is an answering machine until it newly plays back a message from the beginning.

6. The autocall processing method of claim 1, wherein the computer detects an answering machine connection using a detection model that has been trained with one or more types of mechanical sound features, and the mechanical sound features obtained by frequency analysis of the input target person's voice.

7. 2. The method of claim 1, wherein the message is recorded voice data.

8. A computer program that causes a computer to execute the auto-call processing method according to any one of claims 1 to 7.

9. An auto-call processing system for playing back messages by auto-call, comprising: An auto-call is made to the target person, a message is played to the connected target person, and the connection status is determined by acquiring the target person's voice. When the connection state is determined to be an answering machine, the auto-call processing system plays back the message from the beginning.

Images