阅读说明：本技术 一种在自动应答中转接人工的处理方法和装置 (Method and device for processing manual transfer in automatic response ) 是由 韩剑 徐敏 于 2021-07-23 设计创作，主要内容包括：本申请公开了一种在自动应答中转接人工的处理方法和装置,所述方法包括：获取客户的通话语音,其中,所述通话语音是所述客户与自动应答对话过程中保存下来的；获取所述通话语音对应的第一信息,其中,所述第一信息包括以下至少之一：将所述通话语音转成文字后提取的关键词、所述通话语音的物理属性信息；根据所述第一信息判断所述客户是否需要转接人工；在需要转接人工的情况下,将所述客户的通话转接至人工,并至少将所述第一信息中的关键词发送到所述人工所在坐席的显示器上。通过本申请解决了现有技术中在自动应答通话过程中没有关注客户需求所导致的问题,从而能够根据客户的需求自动转向人工,提高了客户体验。(The application discloses a method and a device for processing manual transfer in automatic response, wherein the method comprises the following steps: acquiring a call voice of a client, wherein the call voice is stored in the process of the automatic response conversation of the client; acquiring first information corresponding to the call voice, wherein the first information comprises at least one of the following: converting the call voice into characters, and extracting key words and physical attribute information of the call voice; judging whether the customer needs to transfer the manual work according to the first information; and under the condition that manual work needs to be transferred, transferring the call of the customer to manual work, and at least sending the keywords in the first information to a display of a seat where the manual work is located. Through the method and the device, the problem that the user demand is not concerned about in the automatic answering conversation process in the prior art is solved, so that the manual work can be automatically turned according to the demand of the user, and the user experience is improved.)

1. A method for manually switching in automatic response, comprising:

acquiring a call voice of a client, wherein the call voice is stored in the process of the automatic response conversation of the client;

acquiring first information corresponding to the call voice, wherein the first information comprises at least one of the following: converting the call voice into characters, and extracting key words and physical attribute information of the call voice;

judging whether the customer needs to transfer the manual work according to the first information;

and under the condition that manual work needs to be transferred, transferring the call of the customer to manual work, and at least sending the keywords in the first information to a display of a seat where the manual work is located.

2. The method of claim 1, wherein determining whether the customer needs to switch to a manual task based on the first information comprises:

judging whether the decibel of the call voice of the client is increased by more than a preset value at a preset time point;

in the event of a rise above a predetermined value, it is determined that the customer requires a switching labor.

3. The method of claim 1, wherein determining whether the customer needs to switch to a manual task based on the first information comprises:

judging whether keywords extracted after the conversation voice is converted into characters comprise first type words or second type words, wherein the first type words are emotion words, and the second type words are words which the customer expresses and hopes to be converted into manual processing;

determining that the customer requires a transfer labor if the first type of term or the second type of term is included.

4. The method according to any one of claims 1 to 3, wherein after determining that the customer needs to transfer manual work according to the first information, the method further comprises:

and in the call, playing a prompt statement, wherein the prompt statement is used for prompting the client that the call is switched to manual processing after a preset time.

5. The method of claim 4, wherein after the alert statement is played in the call, the method further comprises:

before the manual access to the call, if the call is ended by the client, storing the recording and the contact way of the client;

and allocating staff for return visit according to the recording and the contact way of the client.

6. A device for manually switching a call in an automatic response, comprising:

the system comprises a first acquisition module, a second acquisition module and a third acquisition module, wherein the first acquisition module is used for acquiring the conversation voice of a client, and the conversation voice is stored in the conversation process of the client and an automatic response;

a second obtaining module, configured to obtain first information corresponding to the call voice, where the first information includes at least one of the following: converting the call voice into characters, and extracting key words and physical attribute information of the call voice;

the judging module is used for judging whether the customer needs to transfer the manual work according to the first information;

and the switching module is used for switching the call of the customer to the manual work under the condition that manual work is required to be switched, and at least sending the key words in the first information to a display of the seat where the manual work is located.

7. The apparatus of claim 6, wherein the determining module is configured to:

judging whether the decibel of the call voice of the client is increased by more than a preset value at a preset time point;

in the event of a rise above a predetermined value, it is determined that the customer requires a switching labor.

8. The apparatus of claim 6, wherein the determining module is configured to:

judging whether keywords extracted after the conversation voice is converted into characters comprise first type words or second type words, wherein the first type words are emotion words, and the second type words are words which the customer expresses and hopes to be converted into manual processing;

determining that the customer requires a transfer labor if the first type of term or the second type of term is included.

9. The apparatus according to any one of claims 6 to 8, further comprising, after determining that the customer needs to transfer manual work according to the first information:

and the playing module is used for playing a prompt statement in the call, wherein the prompt statement is used for prompting the client that the call is switched to manual processing after a preset time.

10. The apparatus of claim 9, further comprising:

the storage module is used for storing the recording and the contact way of the client if the call is ended by the client after the prompt statement is played in the call and before the manual access to the call;

and the distribution module is used for distributing staff to return visits according to the recording and the contact way of the client.

Technical Field

The application relates to the field of conversation, in particular to a method and a device for processing manual switching in automatic answering.

Background

With the increasing cost of manpower, the automatic answering system has the function of replacing the manpower, and can answer questions proposed by the customers according to the requirements of the customers and preset logic.

After answering the customer's question, a playback of the effect is generally made, and the automatic answering system asks whether the question has been solved at the end of the call, and can switch to manual work if the predetermined operation is not performed to solve the question. This transfer process is typically prompted only at the end.

For certain situations, such as after-sales quality consultation, the customer may have an impatient mood with the automated answering system, and the customer may have hung up the phone while the automated answering has not yet proceeded to the stage of asking whether to switch to the manual. This can degrade the customer experience.

The inventor finds that no technology for judging the emotion of a client in the automatic answering process and switching to manual work exists in the prior art.

Disclosure of Invention

The embodiment of the application provides a method and a device for processing manual transfer in automatic answering, so as to at least solve the problem caused by no attention to customer requirements in the automatic answering conversation process in the prior art.

According to one aspect of the application, a method for processing manual switching in automatic answering is provided, and comprises the following steps: acquiring a call voice of a client, wherein the call voice is stored in the process of the automatic response conversation of the client; acquiring first information corresponding to the call voice, wherein the first information comprises at least one of the following: converting the call voice into characters, and extracting key words and physical attribute information of the call voice; judging whether the customer needs to transfer the manual work according to the first information; and under the condition that manual work needs to be transferred, transferring the call of the customer to manual work, and at least sending the keywords in the first information to a display of a seat where the manual work is located.

Further, the step of judging whether the customer needs to transfer the manual work according to the first information comprises: judging whether the decibel of the call voice of the client is increased by more than a preset value at a preset time point; in the event of a rise above a predetermined value, it is determined that the customer requires a switching labor.

Further, the step of judging whether the customer needs to transfer the manual work according to the first information comprises: judging whether keywords extracted after the conversation voice is converted into characters comprise first type words or second type words, wherein the first type words are emotion words, and the second type words are words which the customer expresses and hopes to be converted into manual processing; determining that the customer requires a transfer labor if the first type of term or the second type of term is included.

Further, after determining that the customer needs to transfer manual work according to the first information, the method further includes: and in the call, playing a prompt statement, wherein the prompt statement is used for prompting the client that the call is switched to manual processing after a preset time.

Further, after the prompt statement is played in the call, the method further includes: before the manual access to the call, if the call is ended by the client, storing the recording and the contact way of the client; and allocating staff for return visit according to the recording and the contact way of the client.

According to another aspect of the present application, there is also provided a processing apparatus for switching a manual in an automatic response, including: the system comprises a first acquisition module, a second acquisition module and a third acquisition module, wherein the first acquisition module is used for acquiring the conversation voice of a client, and the conversation voice is stored in the conversation process of the client and an automatic response; a second obtaining module, configured to obtain first information corresponding to the call voice, where the first information includes at least one of the following: converting the call voice into characters, and extracting key words and physical attribute information of the call voice; the judging module is used for judging whether the customer needs to transfer the manual work according to the first information; and the switching module is used for switching the call of the customer to the manual work under the condition that manual work is required to be switched, and at least sending the key words in the first information to a display of the seat where the manual work is located.

Further, the determining module is configured to: judging whether the decibel of the call voice of the client is increased by more than a preset value at a preset time point; in the event of a rise above a predetermined value, it is determined that the customer requires a switching labor.

Further, the determining module is configured to: judging whether keywords extracted after the conversation voice is converted into characters comprise first type words or second type words, wherein the first type words are emotion words, and the second type words are words which the customer expresses and hopes to be converted into manual processing; determining that the customer requires a transfer labor if the first type of term or the second type of term is included.

Further, after the determining that the customer needs to transfer the manual work according to the first information, the method further includes: and the playing module is used for playing a prompt statement in the call, wherein the prompt statement is used for prompting the client that the call is switched to manual processing after a preset time.

Further, still include: the storage module is used for storing the recording and the contact way of the client if the call is ended by the client after the prompt statement is played in the call and before the manual access to the call; and the distribution module is used for distributing staff to return visits according to the recording and the contact way of the client.

In the embodiment of the application, the method comprises the steps of obtaining conversation voice of a client, wherein the conversation voice is saved in the conversation process between the client and an automatic response; acquiring first information corresponding to the call voice, wherein the first information comprises at least one of the following: converting the call voice into characters, and extracting key words and physical attribute information of the call voice; judging whether the customer needs to transfer the manual work according to the first information; and under the condition that manual work needs to be transferred, transferring the call of the customer to manual work, and at least sending the keywords in the first information to a display of a seat where the manual work is located. Through the method and the device, the problem that the user demand is not concerned about in the automatic answering conversation process in the prior art is solved, so that the manual work can be automatically turned according to the demand of the user, and the user experience is improved.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this application, illustrate embodiments of the application and, together with the description, serve to explain the application and are not intended to limit the application. In the drawings:

fig. 1 is a flowchart of a processing method for switching between manual operations in automatic response according to an embodiment of the present application.

Detailed Description

It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict. The present application will be described in detail below with reference to the embodiments with reference to the attached drawings.

It should be noted that the steps illustrated in the flowcharts of the figures may be performed in a computer system such as a set of computer-executable instructions and that, although a logical order is illustrated in the flowcharts, in some cases, the steps illustrated or described may be performed in an order different than presented herein.

In the present embodiment, a processing method for switching a human in an automatic answer is provided, and fig. 1 is a flowchart of a processing method for switching a human (or called a human agent) in an automatic answer according to an embodiment of the present application, as shown in fig. 1, the method includes the following steps:

step S102, obtaining the conversation voice of the client, wherein the conversation voice is saved in the conversation process between the client and the automatic response;

step S104, acquiring first information corresponding to the call voice, wherein the first information comprises at least one of the following: converting the call voice into characters, and extracting key words and physical attribute information of the call voice;

as an alternative, a score value may be set for this call, which defaults to 0. And when the condition that manual work needs to be switched is judged according to the first information, the point value is increased, and when the point value is increased to a preset point, the manual work is determined to be switched.

There are many cases of increasing the score value, several of which are listed below:

first one

Under the condition that keywords extracted after the conversation voice is converted into characters comprise first-type words or second-type words, a first numerical value is increased when the score value of the first-type words is detected once, and a second numerical value is increased when the score value of the second-type words is detected once, wherein the second numerical value is larger than the first numerical value, the first-type words are emotion-type words, and the second-type words are words which the customer expression wishes to be converted into manual processing.

Second kind

Whether the decibel of the call voice of the client is increased by more than a preset value at a preset time point can also be judged; in the event of a rise above a predetermined value, increasing the fractional value by a third value; and if the decibel duration of the call exceeds the preset time, increasing the score value by a fourth value, wherein the fourth value is larger than the third value.

Third kind

It can also be judged whether the speech rate of the call of the client is greater than a predetermined speech rate, and if so, the score value is increased by a fifth numerical value.

The three increasing modes can be used independently or in combination.

When the score value exceeds a predetermined score, for example 10 or 20, then it is determined that the call needs to be forwarded to manual processing.

As another embodiment that can be optionally increased, the setting of the predetermined score may be flexibly set according to the degree of busy of human beings, for example, an idle condition of a human seat may be obtained, where the predetermined score is a first threshold value in a case that the idle rate of the human seat exceeds 10%, and the predetermined score is a second threshold value in a case that the human seat is not idle, where the first threshold value is smaller than the second threshold value.

For another example, the level of the customer may be obtained, where the customer level is used to indicate the importance of the customer, and the customer level is divided into three levels: important, general and unimportant, wherein, for the important customer, the current predetermined score multiplied by 0.8 is the corresponding predetermined score of the important customer. This ensures that important customers switch their manual work in priority.

Step S106, judging whether the customer needs to transfer the manual work according to the first information;

and step S108, under the condition that manual work needs to be transferred, transferring the call of the customer to manual work, and at least sending the keywords in the first information to a display of an agent where the manual work is located.

As an optional added embodiment, after switching to a manual work, recording the communication between the client and the manual work, after the call between the client and the manual work is finished, scoring the conversation between the client and the manual work by using a recording file between the client and the manual work, and determining whether a return visit is needed according to the scoring. The scoring can be divided into two categories, one for need of return visit and one for no need of return visit. And under the condition that the score is that the return visit is needed, allocating staff to return visit according to the recording and the contact way of the client. And when the staff is allocated to return visit, selecting the staff with high allocation priority to return visit, wherein the rate of no need of return visit in the scores obtained by the staff with high priority exceeds 80% and is different from the gender of the client. This gives the customer a better experience.

There are many ways to score, for example, a sound recording file may be converted into text information, and a determination may be made from the text information as to whether a third type of word is included, where the third type of word is used to indicate that the customer is satisfied. And under the condition that the text information comprises the third type words, determining that the score corresponding to the sound recording file is not required to return visit.

Or the physical attribute information in the sound recording can also be acquired, wherein the physical attribute information comprises at least one of the following: the decibel value in the recording is high or low, and the speed of speech in the recording is high or low. And determining whether return visit is needed or not according to the physical attribute information.

A return visit score, which defaults to 0, may also be determined. And when the condition that the return visit is needed once is judged, increasing the return visit score, and when the return visit score is increased to a preset score, determining to carry out the return visit.

For example, there are many cases where the return visit score is increased, several of which are listed below:

and under the condition that the keywords extracted after the conversation voice is converted into the characters comprise the third type of words, increasing a sixth numerical value every time the third type of words are detected to appear.

Whether the decibel rise of the call voice of the client exceeds a preset value can be judged; increasing the seventh value in the event of a rise above a predetermined value; and if the decibel duration time of the call exceeds the preset time length, increasing an eighth value, wherein the eighth value is larger than the seventh value.

It can also be judged whether the speech rate of the call of the client is greater than a predetermined speech rate, and if so, the score value is increased by a ninth value.

The problem caused by the fact that customer requirements are not paid attention to in the automatic answering conversation process in the prior art is solved through the steps, and therefore manual operation can be automatically turned to according to the customer requirements, and customer experience is improved.

There are many ways to determine whether to switch manually, for example, it can be determined whether the decibel of the call voice of the customer rises above a predetermined value at a predetermined time point; in the event of a rise above a predetermined value, it is determined that the customer requires a switching labor. For another example, the determining whether the customer needs to switch to the manual work according to the first information includes: judging whether keywords extracted after the conversation voice is converted into characters comprise first type words or second type words, wherein the first type words are emotion words, and the second type words are words which the customer expresses and hopes to be converted into manual processing; determining that the customer requires a transfer labor if the first type of term or the second type of term is included.

In order to make the customer feel better, after the fact that the customer needs to switch to manual work is judged according to the first information, a prompt statement is played in the call, wherein the prompt statement is used for prompting the customer that the call is switched to manual work after a preset time.

As another preferred processing manner, after the prompt statement is played in the call, the method may further include: before the manual access to the call, if the call is ended by the client, storing the recording and the contact way of the client; and allocating staff for return visit according to the recording and the contact way of the client.

In this embodiment, an electronic device is provided, comprising a memory in which a computer program is stored and a processor configured to run the computer program to perform the method in the above embodiments.

The programs described above may be run on a processor or may also be stored in memory (or referred to as computer-readable media), which includes both non-transitory and non-transitory, removable and non-removable media, that implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.

These computer programs may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks, and corresponding steps may be implemented by different modules.

In this embodiment, there is provided an apparatus called a processing apparatus for switching a manual operation in an automatic response, comprising: the system comprises a first acquisition module, a second acquisition module and a third acquisition module, wherein the first acquisition module is used for acquiring the conversation voice of a client, and the conversation voice is stored in the conversation process of the client and an automatic response; a second obtaining module, configured to obtain first information corresponding to the call voice, where the first information includes at least one of the following: converting the call voice into characters, and extracting key words and physical attribute information of the call voice; the judging module is used for judging whether the customer needs to transfer the manual work according to the first information; and the switching module is used for switching the call of the customer to the manual work under the condition that manual work is required to be switched, and at least sending the key words in the first information to a display of the seat where the manual work is located.

The modules in the apparatus correspond to the steps in the method, and each module is used to implement each step in the method, which has been already described in the method embodiment and is not described herein again.

For example, the determining module is configured to: judging whether the decibel of the call voice of the client is increased by more than a preset value at a preset time point; in the event of a rise above a predetermined value, it is determined that the customer requires a switching labor. Or, the judging module is configured to: judging whether keywords extracted after the conversation voice is converted into characters comprise first type words or second type words, wherein the first type words are emotion words, and the second type words are words which the customer expresses and hopes to be converted into manual processing; determining that the customer requires a transfer labor if the first type of term or the second type of term is included.

For another example, after determining that the customer needs to transfer manual work according to the first information, the method further includes: and the playing module is used for playing a prompt statement in the call, wherein the prompt statement is used for prompting the client that the call is switched to manual processing after a preset time. Optionally, the apparatus may further include: the storage module is used for storing the recording and the contact way of the client if the call is ended by the client after the prompt statement is played in the call and before the manual access to the call; and the distribution module is used for distributing staff to return visits according to the recording and the contact way of the client.

In the embodiment, the audio of the client can be converted into the text, then the text is subjected to semantic analysis, related keywords are found from the text, the requirements of the client are analyzed through the keywords, and if the fact that the manual work needs to be accessed is judged, the client is directly prompted whether the manual work needs to be converted or not. In the event of confirmation by the customer, the customer's call is forwarded to the sale. The customer's conversation with the sale is scored after the transfer to the sale, and the score of the sale in this session is recorded.

The above are merely examples of the present application and are not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.