Processing apparatus, program, and processing method

The system enhances IVR usability by segmenting audio and allowing users to control playback through voice commands, addressing the inconvenience of traditional IVR systems.

JP7948027B2Active Publication Date: 2026-11-11OKI ELECTRIC INDUSTRY CO LTD
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
JP2023025935
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Filing Date
2023-02-22
Publication Date
2026-11-11
Estimated Expiration
2043-02-22

AI Technical Summary

Technical Problem

Conventional IVR systems require customers to listen to all options before a specific one, and cannot skip or clarify options until the end, leading to inconvenience.

Method used

The system divides audio into segments at checkpoints, allowing users to control playback by uttering commands like 'again', 'from the beginning', or 'send' to skip or resume from specific points, using voice recognition to match user input with playback control information.

Benefits of technology

Enables users to easily skip unnecessary options, replay desired segments, or restart from the beginning, improving the usability of automated voice responses.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007948027000001
    Figure 0007948027000001
  • Figure 0007948027000002
    Figure 0007948027000002
  • Figure 0007948027000003
    Figure 0007948027000003
Patent Text Reader

Abstract

To improve the user-friendliness of automated voice response.SOLUTION: A processing apparatus includes: a first processing unit which transmits voice data to a first device to reproduce the voice data, the voice data being divided into multiple voice sections based on one or more check points which are associated with one or more multiple reproduction positions; a second processing unit which stops, when receiving, from the first device, voice information based on a speech of a user who uses the first device, transmitting the voice data to the first device and converts the received voice information to text information; and a third processing unit which determines, when the text information matches reproduction control information on reproduction control of the voice data, a reproduction position from which the voice data is to be reproduced, on the basis of the reproduction control information and check points at which the first device has already completed reproduction, and causes the first processing unit to transmit the voice data from the determined reproduction position, to the first device.SELECTED DRAWING: Figure 1
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to a processing apparatus, a program, and a processing method.

Background Art

[0002] In recent years, many companies have contact center systems for responding to inquiries from customers. In contact center systems, in order to efficiently provide services such as connecting incoming calls from customers to appropriate operators and shortening the response time of operators after connection, before connecting incoming calls from customers to operators, information is collected from customers using an automatic voice response function called IVR (Interactive Voice Response).

[0003] In conventional IVR, information is collected from customers by the customer pressing the number button on the telephone to select a number in response to voice options associated with numbers, such as "Please press 1 in the case of ..., 2 in the case of ..., and 3 in the case of ...". In recent years, along with the development of speech recognition technology, IVR that collects information from customers by having the customer utter a number or the like into the telephone instead of pressing the number button on the telephone to select a number has also been realized (see, for example, Patent Document 1).

[0004] In addition, in any of the IVRs, after all options are played back, the customer can request playback again, so that the customer can re-listen to options that were missed or that the customer wishes to confirm.

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Summary of Invention

Problems to be Solved by the Invention

[0006] However, with conventional IVR systems, even if a customer wanted to hear a specific option, they had to listen to all the other options played before that option until the end. Furthermore, if they missed an option or wanted to ask for clarification, they had to wait until all the available options had finished playing.

[0007] Thus, traditional IVR systems have the problem of being inconvenient for customers because they cannot skip options they don't need to hear or immediately ask for clarification on options.

[0008] This invention takes the above points into consideration and aims to propose a processing device, program, and processing method that can improve the usability of automated voice response. [Means for solving the problem]

[0009] In order to solve these problems, the apparatus of the present invention provides: It includes multiple options for collecting information from the user, corresponding to the playback position immediately following each option. The audio can be divided into multiple parts at checkpoints. Playback data A first processing unit transmits to the first device for playback, and the first device Ra Yu When receiving voice information based on the utterance of the user The transmission of the aforementioned playback data is stopped, A second processing unit that converts the aforementioned audio information into text information, Based on the matching data that associates the valid choice information and playback control information up to each checkpoint for each of the aforementioned playback data, The aforementioned text information The system compares the selected options and playback control information corresponding to the previously played checkpoint. It comprises a third processing unit, The third processing unit selects the corresponding playback data when the text information matches the selection information and repeatedly instructs the first processing unit to send the playback data to the first device until the necessary information is collected. Once the necessary information is collected, it transmits the information to the automatic call distribution function. The aforementioned text information If it matches the playback control information, the playback position is determined based on the playback control information and the played-back checkpoints. The first processing unit is instructed to transmit the playback data from the playback position to the first device.

[0010] Furthermore, in the program of the present invention, the computer is It includes multiple options for collecting information from the user, corresponding to the playback position immediately following each option. The audio can be divided into multiple parts at checkpoints. Playback data A first processing unit transmits to the first device for playback, and the first device Ra Yu When receiving voice information based on the utterance of the user The transmission of the aforementioned playback data is stopped, A second processing unit that converts the aforementioned audio information into text information, Based on the matching data that associates the valid choice information and playback control information up to each checkpoint for each of the aforementioned playback data, The aforementioned text information The system compares the selected options and playback control information corresponding to the previously played checkpoint. function as a third processing unit Furthermore, the third processing unit repeatedly instructs the first processing unit to select the corresponding playback data when the text information matches the selection information and to send the playback data to the first device until the necessary information is collected, and once the necessary information is collected, it transmits the information to the automatic call distribution function. the text information If it matches the playback control information, the playback position is determined based on the playback control information and the played-back checkpoints. cause the first processing unit to transmit the reproduction data from the relevant reproduction position to the first device It works like this to do so.

[0011] Further, the processing method of the present invention is a processing method for a processing apparatus, wherein the processing apparatus It includes multiple options for collecting information from the user, corresponding to the playback position immediately following each option. is divided into a plurality of audio segments at checkpoints Playback data transmitting the same to a first device for reproduction, and upon receiving from the first device Ra Yu audio information based on a user's utterance The transmission of the aforementioned playback data is stopped, converting the audio information into text information, Based on the matching data that associates the valid choice information and playback control information up to each checkpoint for each of the aforementioned playback data, the text information The system compares the selected options and playback control information corresponding to the previously played checkpoint. a step and The process of selecting the corresponding playback data when the text information matches the selection information and sending the playback data to the first device is repeated until the necessary information is collected, and once the necessary information is collected, the information is sent to the automatic call distribution function, the text information If it matches the playback control information, the playback position is determined based on the playback control information and the played-back checkpoints. transmitting the reproduction data from the relevant reproduction position to the first device.

[0012] Thereby, when the user utters, for example, "repeat", it is possible to return to the beginning of the audio segment that was being reproduced immediately before and resume reproduction from that audio segment, when the user utters, for example, "from the beginning", it is possible to immediately resume reproducing the audio data from the beginning, and when the user utters, for example, "skip", it is possible to skip the audio segment that was being reproduced immediately before and start reproduction from the next audio segment, thereby enabling such reproduction control. [Advantageous Effects of Invention]

[0013] Thus, the present invention can realize a processing apparatus, a program, and a processing method that can improve the usability of automatic voice response. [Brief Description of the Drawings]

[0014] [Figure 1] is a diagram showing the configuration of a contact center system. [Figure 2] is a diagram showing the configuration of data handled in the contact center system. [Figure 3]This is a sequence diagram showing the operation of the contact center system. [Figure 4] This is a sequence diagram continuing from the sequence diagram of Figure 3. [Mode for Carrying Out the Invention]

[0015] Hereinafter, the mode for carrying out the invention (hereinafter referred to as the present embodiment) will be described in detail with reference to the drawings.

[0016] [1. Configuration of Contact Center System] First, the configuration of the contact center system 1 according to the present embodiment will be described with reference to Figure 1. This contact center system 1 is a system installed in a contact center of an enterprise or the like, for example, and is a system capable of responding to inquiries by telephone or the like from a customer U20.

[0017] This contact center system 1 is composed of a contact center server 100 and operator communication devices 300 used by operators U10, and the contact center server 100 and the operator communication devices 300 are communicably connected via a local area network 400 such as a LAN (Local Area Network). The contact center server 100 is also communicably connected to a customer communication device 500 owned by a customer U20 via a public network 600 such as an IP (Internet Protocol) network.

[0018] Note that in Figure 1, for the sake of simplicity of description, one operator communication device 300 used by one operator U10 is connected to the contact center server 100, but a plurality of operator communication devices 300 used by a plurality of operators U10 may be connected. Similarly, in Figure 1, one customer communication device 500 owned by one customer U20 is connected to the contact center server 100, but a plurality of customer communication devices 500 owned by a plurality of customers U20 may be connected.

[0019] The operator communication device 300 is a device that provides operator U10 with communication with customer communication device 500 connected by contact center server 100. This operator communication device 300 has an input / output device that responds to incoming calls from customer communication device 500 via contact center server 100 and enables communication with customer U20, and a control function for terminating the call.

[0020] The contact center server 100 is a server device that performs the processing of contact center services provided to customer U20. This contact center server 100 is configured as a computer having a CPU (Central Processing Unit), memory, storage means such as a hard disk, and a network interface, and various functions are realized by the CPU executing programs loaded into memory.

[0021] The contact center server 100 has an IVR (Interactive Voice Response) 200 and an ACD (Automatic Call Distributor) (not shown). The IVR 200 has the function of answering an incoming call from the customer communication device 500, sending a voice question to the customer communication device 500 and receiving a response, and repeating this cycle until the purpose of the call is collected. Furthermore, this IVR 200 is a voice-input compatible IVR that can collect information from customer U20 by receiving the customer U20's voice as a response from the customer communication device 500.

[0022] Furthermore, the contact center server 100 has the function of sending the purpose of the call, collected by the IVR 200, to the ACD.

[0023] Specifically, the IVR200 includes a voice receiving unit 210, a voice recognition unit 220, a voice transmission unit 230, a control unit 240, a playback data storage unit 250, and a matching data storage unit 260.

[0024] The voice receiving unit 210 receives the voice of customer U20 transmitted from the customer communication device 500 to the contact center server 100 and sends it sequentially to the voice recognition unit 220. The voice recognition unit 220 sequentially converts the voice sent from the voice receiving unit 210 (i.e., the voice of customer U20) into text information and sends it to the control unit 240. The voice transmitting unit 230 receives the voice data (playback data described later) from the control unit 240 for the contact center server 100 to transmit to the customer communication device 500 and transmits it to the customer communication device 500.

[0025] The playback data storage unit 250 stores multiple playback data sets that have been prepared in advance for customer U20 to listen to. Each of the stored playback data sets is assigned delimiter information (called checkpoints) for each meaningful section. The matching data storage unit 260 stores matching data that defines which playback data stored in the playback data storage unit 250 should be played and how, in response to an incoming call from customer U20.

[0026] Here, Figure 2(A) shows an example of playback data stored in the playback data storage unit 250, and Figure 2(B) shows an example of matching data stored in the matching data storage unit 260.

[0027] The playback data storage unit 250 stores playback data x which contains the message, "If it is ..., say 1; if it is ..., say 2; if it is ..., say 6." In other words, this playback data x is audio data for collecting the purpose of the call from customer U20, and consists of multiple options associated with numbers, such as "If it is ..., say 1," and commands prompting customer U20 to speak, such as "Say 6."

[0028] Furthermore, in this playback data x, checkpoints are attached immediately after each option, such as checkpoint cp-x1 immediately after the part "...if...is 1". In other words, playback data x is divided into multiple audio parts, with one or more checkpoints associated with one or more playback positions (playback positions immediately following an option), and these checkpoints serve as boundaries. These checkpoints are information used to manage the playback of playback data x on an option-by-option basis.

[0029] Although not shown in Figure 2, the playback data storage unit 250 also stores various other playback data, including playback data x, for collecting the purpose of the call from customer U20.

[0030] On the other hand, the matching data storage unit 260 stores matching data for each incoming call (specifically, the incoming telephone number), which associates the playback data to be played back for that incoming call with the valid options (specifically, the option numbers) and valid playback control words (hereinafter referred to as playback control words) for each checkpoint attached to the playback data.

[0031] For example, the matching data indicates that the playback data to be played back for an incoming call (incoming phone number) X is playback data x, and that the number of the valid option up to checkpoint cp-x1 (i.e., the number of the option before checkpoint cp-x1) attached to playback data x is "1", and the valid playback control words are "again", "from the beginning", and "send", and that the number of the valid options up to checkpoint cp-x2 is "1" and "2", and the valid playback control words are "again", "from the beginning", and "send", and so on.

[0032] The matching data also indicates that the playback data to be played back for the incoming call (incoming phone number) Y is playback data y, that the valid option number up to checkpoint cp-y1 attached to playback data y is "1" and the valid playback control words are "again", "from the beginning", and "send", and that the valid option numbers up to checkpoint cp-y2 are "1" and "2" and the valid playback control words are "again", "from the beginning", and "send", etc.

[0033] Returning to Figure 1, the control unit 240, upon receiving a call from customer U20, refers to the matching data storage unit 260 to identify the playback data corresponding to the incoming phone number (i.e., the phone number dialed by customer U20), retrieves the identified playback data from the playback data storage unit 250, and transmits it to the voice transmission unit 230. The control unit 240 also has a function to manage up to which checkpoint the playback data has been played, and, upon receiving text information indicating customer U20's voice from the voice recognition unit 220, interrupts the transmission of the playback data and compares the received text information with the valid options and playback control words defined in the matching data up to the most recent checkpoint.

[0034] Furthermore, the control unit 240 has the function of sending playback data to the voice transmission unit 230 based on the matching result, and the function of sending the information (i.e., the purpose of the call) to the ACD when the collection of necessary information from customer U20 is complete (i.e., the collection of the purpose of the call is complete). The configuration of the contact center system 1 is as described above.

[0035] [2. Operation of the Contact Center System] Next, the operation of the contact center system 1 will be described. Here, we will describe the operation of the IVR 200 of the contact center server 100 when it provides an automated voice response to an incoming call from customer U20. Figures 3 and 4 show the sequence chart of this operation. Note that this operation is mainly performed by the control unit 240 of the IVR 200.

[0036] First, suppose a call is received from a customer communication device 500 of a certain customer U20 to, for example, an incoming telephone number X (SP1). At this point, the control unit 240 identifies the playback data corresponding to the incoming telephone number X as playback data x by referring to the matching data stored in the matching data storage unit 260 based on the incoming telephone number X (SP2), and retrieves the playback data x from the playback data storage unit 250 (SP3).

[0037] The control unit 240 then sends the acquired playback data x to the voice transmission unit 230, which then transmits it to the customer communication device 500. This allows customer U20 to hear the audio based on the playback data x (i.e., the options that constitute the questions and answers to customer U20) and prompts them to make a voice input (i.e., select an option) (SP5). At this time, the control unit 240 sends the playback data x to the voice transmission unit 230 from the beginning, and the playback data x is played back from the beginning on the customer communication device 500 side.

[0038] Here, let's assume that customer U20 speaks after passing checkpoint cp-x2 of the playback data x but before reaching the next checkpoint cp-x3 (i.e., while listening to the third option) (SP6). At this time, customer U20's speech is transmitted as voice data from the customer communication device 500 to the contact center server 100, received by the voice receiving unit 210, and sent from the voice receiving unit 210 to the voice recognition unit 220 for text conversion (SP7). Then, the text information of customer U20's speech is sent to the control unit 240 (SP8).

[0039] The control unit 240 then temporarily suspends sending the playback data x to the audio transmission unit 230 (SP9), and compares the valid choice numbers ("1", "2") and valid playback control words ("again", "from the beginning", "send") described in the matching data up to checkpoint cp-x2 of the playback data x with the text information received from the speech recognition unit 220 (i.e., the speech of customer U20) (SP10).

[0040] If the text information matches option number "1" or "2" at this point (SP101), the control unit 240 determines that it has finished collecting the purpose of the call because customer U20 has selected an option, terminates sending the playback data x, sends the collected purpose of the call (i.e., the option selected by customer U20) to the ACD function (SP102), and proceeds to operator assistance.

[0041] Furthermore, if the text information matches the playback control word "from the beginning" (SP111 in Figure 4), the control unit 240 decides to play the playback data x from the beginning and resends the playback data x to the voice transmission unit 230 from the beginning (SP112 → SP4). As a result, the playback data x is played back from the beginning on the customer communication device 500 side. In this case, since customer U20 has already listened up to checkpoint cp-x2 (i.e., up to the second option), when the control unit 240 plays the playback data x back from the beginning, even if customer U20 speaks before reaching checkpoint cp-x2, the valid option numbers and valid playback control words will be limited to up to checkpoint cp-x2. In other words, when playback data x is played back, the option numbers that customer U20 has listened to up to that point (e.g., "1" and "2") and the playback control words associated with those numbers (e.g., "again", "from the beginning", "send") become valid.

[0042] Furthermore, if the text information matches the playback control word "again" (SP121 in Figure 4), the control unit 240 decides to play the playback data x from the previous checkpoint cp-x2 and resends the playback data x from the previous checkpoint cp-x2 to the voice transmission unit 230 (SP122 → SP4). As a result, the playback data x is played back again on the customer communication device 500 side from the previous checkpoint cp-x2 (i.e., from the second option).

[0043] Furthermore, if the text information matches the playback control word "send" (SP131 in Figure 4), the control unit 240 decides to play the playback data x from the next checkpoint cp-x3 and sends the playback data x to the audio transmission unit 230 from the next checkpoint cp-x3 (SP132 → SP4). As a result, the playback data x is played back from the next checkpoint cp-x3 on the customer communication device 500 side (i.e., playback of the third option is skipped and playback starts from the fourth option). In this case, the control unit 240 treats customer U20 as having listened up to checkpoint cp-x3. In other words, if playback data x is skipped, the number of the skipped option (e.g., "3") and the playback control word associated with that number (e.g., "again", "from the beginning", "send") also become valid.

[0044] On the other hand, suppose customer U20 does not speak even after listening to the playback data x to the end. In this case, the control unit 240 resends the playback data x to the voice transmission unit 230 from the beginning until customer U20 speaks. As a result, on the customer communication device 500 side, the playback data x is played back from the beginning until customer U20 speaks.

[0045] In this manner, when the playback data x is being repeatedly played back from the beginning, if customer U20 speaks, the control unit 240 treats it as if customer U20 has listened to the playback data x to the end at least once, and therefore has listened up to the last checkpoint cp-x6. In other words, after playing the playback data x to the end, if it is played back again from the beginning, all the option numbers (e.g., "1", "2", "3", "4", "5", "6") and the playback control words associated with those numbers (e.g., "again", "from the beginning", "send") become active. The operation when the contact center system 1 performs automated voice response is as described above.

[0046] [3. Summary and Effects] As described above, the contact center system 1 of this embodiment includes a contact center server 100 as a processing unit and computer, and an operator communication device 300.

[0047] The contact center server 100 also has a first processing unit (voice transmission unit 230 and control unit 240) that transmits playback data, which is audio data in which one or more checkpoints are associated with one or more playback positions and which is divided into multiple audio parts at the checkpoints, to the customer communication device 500, which is the first device, for playback.

[0048] Furthermore, when the contact center server 100 receives voice data from the customer communication device 500 as voice information based on the speech of customer U20 as a user of the customer communication device 500, it has a second processing unit (voice receiving unit 210, voice recognition unit 220, voice transmission unit 230, and control unit 240) that stops transmitting playback data to the customer communication device 500 and converts the received voice data into text information.

[0049] Furthermore, the contact center server 100 has a third processing unit (control unit 240, matching data storage unit 260) which, when text information matches a playback control word as playback control information related to playback control of playback data, determines from which playback position the playback data should be transmitted to the customer communication device 500 for playback based on the playback control word and checkpoints that have been played back by the customer communication device 500 up to the present, and causes the first processing unit to transmit the playback data to the customer communication device 500 from the determined playback position.

[0050] In other words, the third processing unit of the contact center server 100 decides to play the playback data from the previous checkpoint if the text information indicating the customer U20's speech matches a playback control word, such as "again," which indicates that the playback data should be returned to the previous checkpoint.

[0051] Furthermore, the third processing unit of the contact center server 100 decides to play the playback data from the next checkpoint if the text information indicating the customer U20's speech matches a playback control word, such as "send," which indicates that the playback data should be advanced to the next checkpoint.

[0052] Furthermore, the third processing unit of the contact center server 100 decides to play the playback data from the beginning if the text information representing the customer U20's speech matches a playback control word, such as "from the beginning," which indicates that the playback data should be played from the beginning.

[0053] As a result, the contact center server 100 can perform playback control in response to voice input from customer U20, such as returning to the beginning of the previously played audio portion (choice) (i.e., the previous checkpoint) and replaying it from that audio portion, immediately replaying the playback data from the beginning, or skipping the previously played audio portion (choice) and playing from the beginning of the next audio portion (i.e., the next checkpoint). Thus, the contact center server 100 of this embodiment can improve the usability of the automated voice response.

[0054] Furthermore, since the contact center server 100 has an IVR 200 (voice receiving unit 210, voice recognition unit 220, voice transmission unit 230, and control unit 240) that responds to voice input from customer U20, customer U20 can easily and quickly listen to the desired options again, listen to the audio again from the beginning, or skip options that do not need to be listened to, by speaking (i.e., voice input) corresponding to the playback control words while listening to the audio of the playback data.

[0055] [4. Other Embodiments] [4-1. Other Embodiments 1] In the above-described embodiment, the control unit 240 of the contact center server 100 functions as the first processing unit, second processing unit, and third processing unit described above. However, the invention is not limited to this configuration. For example, instead of the control unit 240, the contact center server 100 may be provided with a first control unit that functions as the first processing unit, a second control unit that functions as the second processing unit, and a third control unit that functions as the third processing unit.

[0056] [4-2. Other Embodiments 2] In the above-described embodiment, the playback data, which is audio data, is stored in the playback data storage unit 250, and the playback data is transmitted from the audio transmission unit 230 to the customer communication device 500 for playback. However, the invention is not limited to this, and the playback data storage unit 250 may store the playback data, which is text data, and the playback data, which is text data, may be converted into playback data, which is then transmitted from the audio transmission unit 230 to the customer communication device 500 for playback. In this case, for example, an audio synthesis unit may be provided before the audio transmission unit 230, and the audio synthesis unit may be used to convert the playback data into text audio.

[0057] [4-3. Other Embodiments 3] Furthermore, in the embodiment described above, the matching data storage unit 260 stores, for each incoming call (specifically, the incoming telephone number), matching data that associates playback data to be played back for that incoming call with the valid choice numbers and valid playback control words for each checkpoint attached to the playback data.

[0058] In this matching data, options are identified using numbers such as "1" and "2," but this is not the only option. You can also use words that identify the options instead of numbers (for example, "..." from "...in the case of 1").

[0059] [4-4. Other Embodiments 4] Furthermore, in the embodiment described above, "again," "from the beginning," and "send" were used as playback control words, but the system is not limited to these; any other words that have meaning related to playback control may be used. For example, words such as "repeat," "from the beginning," and "skip" may be used. In addition, for each playback control word, multiple words that are likely to be spoken by the customer U20 may be associated with it and stored as matching data in the matching data storage unit 260. For example, for the playback control word "send," words such as "send," "skip," "skip," and "skipped" may be associated with it. This allows for flexible responses to the customer U20's speech. Furthermore, an explanation of such playback control words may be included at the beginning of the playback data (before the options) to notify the customer U20 about the playback control words.

[0060] [4-5. Other Embodiments 5] Furthermore, although the above-described embodiment applied the present invention to a contact center system 1 that conducts calls between operator U10 and customer U20, the present invention can be applied to various contact center systems that have an automated voice response function.

[0061] Furthermore, in the above-described embodiment, a contact center server 100 was used as a specific example of the processing unit and computer. However, the invention is not limited to this, and a processing unit and computer with a partially different configuration from the contact center server 100 may be used as long as they function in the same way as the contact center server 100.

[0062] [4-6. Other Embodiments 6] Furthermore, the present invention is not limited to the embodiments described above. That is, the scope of application of the present invention extends to embodiments that arbitrarily combine some or all of the embodiments described above, or embodiments that extract some of them.

[0063] [5. Addendum] With regard to the embodiments described above, the following additional information is disclosed.

[0064] To solve the above problems, the processing apparatus of the present invention includes: a first processing unit that transmits to a first device and plays back audio data, in which one or more checkpoints are associated with one or more playback positions and which are divided into multiple audio parts at the checkpoints; a second processing unit that, upon receiving audio information from the first device based on the utterance of a user using the first device, stops transmitting the audio data to the first device and converts the received audio information into text information; and a third processing unit that, when the text information matches playback control information relating to the playback control of the audio data, determines from which playback position the audio data should be played back based on the playback control information and the checkpoints that have been played back by the first device so far, and causes the first processing unit to transmit the audio data to the first device from the determined playback position.

[0065] Furthermore, in the present invention, the third processing unit determines to play the audio data from the previous checkpoint if the text information matches the playback control information indicating that the audio data should be returned to the previous checkpoint.

[0066] Furthermore, in the present invention, the third processing unit determines to play the audio data from the next checkpoint if the text information matches the playback control information indicating that the audio data should be advanced to the next checkpoint.

[0067] Furthermore, in the present invention, the third processing unit determines to play the audio data from the beginning if the text information matches the playback control information indicating that the audio data should be played from the beginning.

[0068] Furthermore, in the present invention, the voice data is voice data for an automated voice response to allow the user to select a desired option, and is divided into multiple voice portions that represent multiple options at the checkpoint, and the third processing unit terminates the transmission of the voice data by the first processing unit when the text information matches one of the multiple options included in the voice data.

[0069] Furthermore, in the program of the present invention, the computer functions as a first processing unit that transmits audio data, in which one or more checkpoints are associated with one or more playback positions and which is divided into multiple audio parts at the checkpoints, to a first device for playback; a second processing unit that, upon receiving audio information from the first device based on the utterance of a user using the first device, stops transmitting the audio data to the first device and converts the received audio information into text information; and a third processing unit that, when the text information matches playback control information relating to the playback control of the audio data, determines from which playback position the audio data should be played back based on the playback control information and the checkpoints that have been played back by the first device so far, and causes the first processing unit to transmit the audio data to the first device from the determined playback position.

[0070] Furthermore, the processing method of the present invention is a processing method for a processing device, comprising the steps of: transmitting audio data, in which one or more checkpoints are associated with one or more playback positions and which is divided into multiple audio parts at the checkpoints, to a first device for playback; receiving audio information from the first device based on the utterance of a user using the first device, stopping the transmission of the audio data to the first device and converting the received audio information into text information; and, if the text information matches playback control information relating to the playback control of the audio data, determining from which playback position the audio data should be played back based on the playback control information and the checkpoints that have been played back by the first device to date, and transmitting the audio data to the first device from the determined playback position.

[0071] The system allows for playback control such as, for example, the user saying "one more time" to return to the beginning of the previously played audio section and restart playback from that section; for example, saying "start from the beginning" to immediately restart playback from the beginning of the audio data; or for example, saying "skip" to skip the previously played audio section and start playback from the next audio section. Thus, the present invention can realize a processing device, program, and processing method that can improve the usability of automatic voice response. [Industrial applicability]

[0072] This invention can be widely used in customer service systems, such as contact center systems. [Explanation of symbols]

[0073] 1... Contact center system, 100... Contact center server, 200... IVR, 210... Voice receiving unit, 220... Voice recognition unit, 230... Voice transmitting unit, 240... Control unit, 250... Playback data storage unit, 260... Matching data storage unit, 300... Operator communication device, 400... Private network, 500... Customer communication device, 600... Public network, U10... Operator, U20... Customer.

Claims

1. A processing apparatus, A first processing unit transmits playback data to a first device for playback, which includes multiple options for collecting information from the user and is divided into multiple audio parts at checkpoints corresponding to the playback position immediately following each option. A second processing unit, upon receiving voice information based on the user's voice from the first device, stops transmitting the playback data and converts the voice information into text information. The system includes a third processing unit that, based on matching data which associates the selection information and playback control information valid up to each checkpoint for each of the aforementioned playback data, matches the text information with the selection information and playback control information corresponding to the playback checkpoint. The third processing unit selects the corresponding playback data when the text information matches the selection information and repeatedly causes the first processing unit to transmit the playback data to the first device until the necessary information is collected. Once the necessary information is collected, it transmits the information to the automatic call distribution function. When the text information matches the playback control information, it determines the playback position based on the playback control information and the playback checkpoints, and causes the first processing unit to transmit the playback data from the playback position to the first device.

2. The third processing unit is, If the text information matches the playback control information indicating that the playback data should be returned to the previous checkpoint, it is decided to play the playback data from the previous checkpoint. The apparatus according to claim 1.

3. The third processing unit is, If the text information matches the playback control information indicating that the playback data should be advanced to the next checkpoint, it is decided to play the playback data from the next checkpoint. The apparatus according to claim 1.

4. The third processing unit is, If the text information matches the playback control information indicating that the playback data should be played from the beginning, it is decided to play the playback data from the beginning. The apparatus according to claim 1.

5. Computers A first processing unit transmits playback data to a first device for playback, which includes multiple options for collecting information from the user and is divided into multiple audio parts at checkpoints corresponding to the playback position immediately following each option. A second processing unit, upon receiving voice information based on the user's voice from the first device, stops transmitting the playback data and converts the voice information into text information. Based on matching data that associates the valid choice information and playback control information up to each checkpoint for each of the aforementioned playback data, the third processing unit functions to match the text information with the choice information and playback control information corresponding to the playback checkpoint. Furthermore, a program to operate the third processing unit to select the corresponding playback data when the text information matches the selection information and to have the first processing unit transmit the playback data to the first device, repeating this process until the necessary information is collected, and once the necessary information is collected, to transmit the information to the automatic call distribution function, and when the text information matches the playback control information, to determine the playback position based on the playback control information and the playback checkpoint, and to have the first processing unit transmit the playback data from that playback position to the first device.

6. A processing method for a processing apparatus, wherein the processing apparatus is The process involves sending playback data, which includes multiple options for collecting information from the user and is divided into multiple audio sections based on checkpoints corresponding to the playback position immediately following each option, to a first device for playback; The first device receives voice information based on the user's voice, stops transmitting the playback data, and converts the voice information into text information. The steps include: comparing the text information with the selected information and playback control information corresponding to the played-through checkpoints, based on matching data that associates the selected information and playback control information valid up to each checkpoint for each of the playback data; The process of selecting the corresponding playback data when the text information matches the selection information and transmitting the playback data to the first device is repeated until the necessary information is collected, and once the necessary information is collected, the information is transmitted to the automatic call distribution function. If the text information matches the playback control information, the playback position is determined based on the playback control information and the playback checkpoints, and the playback data is transmitted from that playback position to the first device. A processing method characterized by having the following:

Citation Information

Patent Citations

  • Voice input and response device

    JP1993136892A

  • Speech recognition and response device

    JP1996265440A

  • Voice automatic response apparatus and method

    JP2019153835A

  • Interactive system, interactive device and interactive program

    JP2020076869A

  • Voice barge-in in telephony speech recognition

    WO2002052546A1