WO2014173342A1

WO2014173342A1 - Method and device for identifying automatic response of calling system

Info

Publication number: WO2014173342A1
Application number: PCT/CN2014/077731
Authority: WO
Inventors: 张伟; 刘澍; 张武雄
Original assignee: 中兴通讯股份有限公司
Priority date: 2013-08-30
Filing date: 2014-05-16
Publication date: 2014-10-30
Also published as: CN104427076A

Abstract

Disclosed are a method and device for identifying an automatic response of a calling system. In the present invention, the identification server receives a response audio data packet transmitted by a response device of a calling system by means of an audio communication channel established with the response device of the calling system; the identification server analyzes the response audio data packet so as to determine whether the response audio data packet satisfies a pre-set automatic response parameter; when it is determined that the response audio data packet satisfies the pre-set automatic response parameter, the identification server determines that the voice response type corresponding to the response audio data packet is automatic response.

Description

Calling system automatic answering identification method and device

The present invention relates to the field of communications, and in particular, to an identification method and apparatus for automatic answering of a call system. Background technique

The NGCC (Next Generation Call Center) call system has broad market prospects in the outsourcing industry and overseas. The call system generally responds with automatic voice response through the answering device.

The call system includes a control device that controls a business process in the call system, the call system further includes an identification server that provides media processing functions in the basic and enhanced services for all audio and video related media processing, including video and Audio RTP (Real Time Transport Protocol) The conversion of data streams to video and audio files. At the same time, it is also responsible for receiving the user's DTMF (Dual Tone Multi-Frequency) input through the terminal, guiding the voice of the broadcast service, and displaying a dynamic boot screen. It has a SIP (Text-Based Protocol) protocol and MSML (Media Session Markup Language and Media Object Markup Language) MOML capabilities that enable it to interact with users throughout the session process under the control of the application server APP.

The identification server can recognize the called response, the called busy, and no response. For the called response, there is a certain proportion of automatic answering devices, including, for example, modems, faxes, telephone messages, voice mail, and secretarial desks that need to recognize the automatic answering device response. However, existing identification servers are unable to identify the portion of the called response that is automatically answered. Summary of the invention

In order to solve the existing technical problems, the embodiments of the present invention mainly provide a method and device for identifying an automatic response of a call system.

An embodiment of the present invention provides a method for identifying an automatic response of a call system, the method comprising: when receiving a response identification request sent by a response device in a call system, the identification server establishes an audio communication channel with the response device;

And the recognition server receives the response audio data packet sent by the response device based on the established audio communication channel;

The identification server analyzes the response audio data packet to determine whether the response audio data packet satisfies a preset automatic response parameter, and when determining that the response audio data packet satisfies a preset automatic response parameter, the identification server determines the response The voice response type corresponding to the audio data packet is an automatic response.

The present invention also provides an identification device for automatically answering a call system, the device comprising: a data transmission module configured to: when receiving a response identification request sent by a response device in the call system, the identification server establishes with the response device Audio communication channel; and

Receiving a response audio data packet sent by the response device based on the established audio communication channel;

An identification module, configured to analyze the response audio data packet, to determine whether the response audio data packet meets a preset automatic response parameter, and when determining that the response audio data packet meets a preset automatic response parameter, determining the The voice response type corresponding to the response audio data packet is an automatic response.

Compared with the prior art, in the embodiment of the present invention, the identification server receives the response audio data packet sent by the response device by establishing an audio communication channel with the answering device in the calling system; and the identification server analyzes the response audio data packet to determine Whether the response audio data packet satisfies a preset automatic response parameter, and determines that the response audio data packet satisfies a preset automatic response When the parameter is, the identification server determines that the voice response type corresponding to the response audio data packet is an automatic response. The realization realizes the automatic response in the called response, so that the control device in the calling system controls the business process of the calling system according to the recognition result. DRAWINGS

1 is a specific flowchart of a first embodiment of a method for identifying an automatic response of a call system according to the present invention;

2 is a specific flowchart of a second embodiment of a method for identifying an automatic response of a call system according to the present invention;

3 is a specific flowchart of a third embodiment of a method for identifying an automatic response of a call system according to the present invention;

4 is a specific flowchart of a fourth embodiment of a method for identifying an automatic response of a call system according to the present invention;

FIG. 5 is a specific flowchart of a first embodiment of an apparatus for identifying an automatic answering system of a call system according to the present invention; FIG.

Fig. 6 is a detailed flow chart showing a second embodiment of the apparatus for automatically answering the call system of the present invention.

The implementation, functional features and advantages of the objects of the present invention will be further described in conjunction with the embodiments herein. detailed description

It is understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

As shown in FIG. 1, it is a specific flowchart of the first embodiment of the method for identifying the automatic response of the call system of the present invention.

It should be emphasized that the flowchart shown in FIG. 1 is only a preferred embodiment, and the technology in the field It is to be understood that any embodiment constructed around the inventive concept should not depart from the scope of the following technical solutions:

Receiving, by the answering device, a response identification request sent by the answering device, the identification server establishes an audio communication channel with the answering device; and based on the established audio communication channel, the identifying server receives the answering audio data packet sent by the answering device The identification server analyzes the response audio data packet to determine whether the response audio data packet satisfies a preset automatic response parameter, and when determining that the response audio data packet satisfies a preset automatic response parameter, the identification server determines the The voice response type corresponding to the response audio data packet is an automatic response.

The following is a specific step in the embodiment to gradually recognize the automatic response of the answering device in the calling system:

Step S11: When receiving the response identification request sent by the answering device in the calling system, the identification server establishes an audio communication channel with the answering device.

Specifically, the call flow of the call system is started, the answering device in the call system receives the media from the phone user, and sends a called response audio data packet based on the received media, and the answering device sends a response identification request to the identification server, when Upon receiving the response identification request, the identification server establishes an audio communication channel with the answering device. The answering device in the calling system includes at least one communication channel for receiving media from a telephone user; the communication channel for receiving media from the telephone user corresponds to a communication channel provided with a response audio data packet; Establishing an audio communication channel with the response device; for example, when the response device receives the media of the telephone user through the A communication channel, the response device transmits the called response audio data packet based on the received media, The communication channel sent by the response audio data packet is a B communication channel set with the A communication channel mapping; the response device sends a response identification request to the identification server, and when receiving the response identification request, the identification server establishes the response device Audio communication channel of the B communication channel; as described above, the identification server is responsive to the number of communication channels that the response device answers to the audio data packet - corresponding to the establishment of the response device Audio communication channel.

Step S12: Based on the established audio communication channel, the identification server receives the response audio data packet sent by the response device.

Step S13: The identification server analyzes the response audio data packet to determine whether the response audio data packet satisfies a preset automatic response parameter.

Step S14: When it is determined that the response audio data packet meets the preset automatic response parameter, the identification server determines that the voice response type corresponding to the response audio data packet is an automatic response.

Specifically, based on an audio communication channel between the established identification server and the response device, the identification server receives the response audio data packet sent by the response device, and the identification server analyzes the response audio data packet to determine the response. Whether the audio data packet meets the preset automatic response parameter, and when determining that the response audio data packet meets the preset automatic response parameter, the identification server determines that the voice response type corresponding to the response audio data packet is an automatic response; When the response audio data packet does not satisfy the preset automatic response parameter, the identification server determines that the voice response type corresponding to the response audio data packet is a manual response. In this embodiment, the preset automatic response parameter may be a preset mute reference ratio, or may be a preset continuous mute reference duration or a preset continuous speech reference duration, and the like. In the other embodiments of the present invention, the preset automatic response parameter may also be a mute reference ratio or a continuous mute reference duration or a preset time period. The continuous voice reference duration, etc., the preset time may be an applicable duration set by a user such as 30s or 40s in advance. For example, an audio communication channel C is established between the identification server and the response communication channel B of the response device, and the identification server receives the response audio data packet D from the response communication channel B in the C audio communication channel receiving call system, the pre- The set auto-answer parameter takes the preset mute reference ratio as an example. The preset mute reference ratio is 25%. The recognition server analyzes the response audio data packet D to obtain the mute ratio in the response audio data packet D. If the obtained mute ratio in the response audio packet D is 20%, the obtained mute ratio is 20%. The recognition server determines that the response audio data packet D satisfies the preset automatic response parameter, that is, the recognition server determines that the voice response type corresponding to the response audio data packet of the silence ratio is 20% is automatic. Answering; if the obtained mute ratio in the response audio data packet D is 30%, the obtained mute ratio 30% is greater than the preset mute reference ratio 25%, and the identification server determines that the response audio data packet D does not satisfy the preset automatic answering parameters, i.e., identifying the other server ¹ J determines the ratio of 30% mute audio packet corresponding to the response voice response type manual answer. The identification server sends the recognition result of the voice response type to the automatic response to the control device in the call system, so that the control device controls the business process of the call system according to the recognition result.

The identification server receives the response audio data packet sent by the response device by establishing an audio communication channel with the answering device in the calling system; the identification server analyzes the response audio data packet to determine whether the response audio data packet satisfies The automatic answering parameter is configured to: when determining that the response audio data packet meets the preset automatic response parameter, the identification server determines that the voice response type corresponding to the response audio data packet is an automatic response. The realization recognizes the automatic response in the called response, so that the control device in the calling system controls the business process of the calling system according to the recognition result.

As shown in FIG. 2, it is a specific flowchart of the second embodiment of the method for identifying the automatic response of the call system of the present invention.

Based on the first embodiment, after step S14, the method further includes:

Step S15: The identification server closes the audio communication channel, and performs deletion detection on the saved data, and deletes data that meets the preset deletion condition.

Specifically, after determining that the response audio data packet meets the preset automatic response parameter, after the identification server determines that the voice response type corresponding to the response audio data packet is an automatic response, the identification server closes the response device. The established audio communication channel, and the deleted data is deleted and detected, and the data that meets the preset deletion condition is deleted. To identify the server Taking the audio communication channel C established between the response communication channel B of the response device as an example, the response audio data packet D transmitted by the response device is received through the audio communication channel C, and the response audio data packet D is determined. After the automatic response parameter is met, the identification server closes the audio channel channel C; the preset deletion condition may be a reference storage duration for saving data, or may be other parameters for saving data set by the user in advance, the preset The deletion condition is taken as an example of the storage duration of the saved data. If the storage duration of the saved data is detected, if the storage duration of the saved data is greater than the reference storage duration, the found saved data is deleted, and the reference is deleted. The save time can be 10 days or 15 days, or it can be other reference save time set by the user in advance. In another embodiment of the present invention, when the preset time is reached, the identification server performs deletion detection on the saved data, and deletes data that meets the preset deletion condition, and is not limited to when the communication channel is closed. The server performs the deletion detection on the saved data, and the preset time may be 1 day or 5 days, or may be any other time interval or time point set by the user in advance.

After receiving and recognizing the response audio data packet sent by the response device, the established communication channel is closed, P strives to lower the running load of the identification server, improves the running speed, and deletes the saved data of the identification server, and utilizes the identification reasonably The storage space of the server increases the processing speed.

As shown in FIG. 3, it is a specific flowchart of the third embodiment of the method for identifying the automatic response of the call system of the present invention.

Based on the foregoing first embodiment, when the preset automatic response parameter is the preset continuous mute reference duration, step S13 further includes:

Step S16: Identify the continuous mute duration of the audio obtained by the server from the response audio data packet. Step S17: When the acquired continuous mute duration is less than the preset continuous mute reference duration, the identification server determines that the response audio data packet satisfies the preset auto-answer parameter.

Specifically, the preset automatic response parameter is a preset continuous silent reference duration, The preset continuous mute reference duration can be set to 0.7s or 1.2s, or it can be the time interval value obtained by any other user through actual detection. Take the preset continuous mute reference duration of 0.7s as an example, identify the server analysis. Receiving the response audio data packet sent by the response device, and obtaining the longest continuous silent duration from the received response audio data packet, and if the longest continuous silent duration obtained is 0.6s, the longest acquisition time is obtained. The continuous mute duration is 0.6s less than the preset continuous mute reference duration of 0.7s, and the recognition server determines that the longest continuous mute duration of 0.6s response audio data packet satisfies the preset auto-answer parameter, that is, the identification server determines the longest continuous The voice response type corresponding to the response audio packet with a silence duration of 0.6s is an automatic response. If the longest continuous silence duration is 1.6s, the obtained continuous silence duration is 1.6s longer than the continuous silent reference duration of 0.7s. The server determines that the response audio packet with the longest continuous silent duration of 1.6s does not satisfy the preset automatic response parameter, that is, the identification service Determines the length of the audio 1.6s response packet corresponding to the type of artificial response voice response longest continuous silence. In other embodiments of the present invention, the preset automatic response parameter may also be a preset continuous mute reference duration within a preset time period.

Obtaining, by analyzing the response audio data packet, a longest continuous silent duration from the received response audio data packet, where the recognition server determines the response audio when the acquired continuous silent duration is less than a preset continuous silent reference duration The data packet satisfies the preset automatic response parameter, and determines that the voice response type corresponding to the response audio data packet is an automatic response. The realization recognizes the automatic response in the called response, so that the control device in the calling system controls the business process of the calling system according to the recognition result.

As shown in FIG. 4, it is a specific flowchart of a fourth embodiment of the method for identifying an automatic answering system of the present invention.

Based on the foregoing first embodiment, when the automatic response parameter is a continuous voice reference duration, step S13 further includes:

Step S18: Identify a continuous voice duration that the server obtains audio from the response audio data packet; Step S19: When the acquired continuous voice duration is greater than the preset continuous voice reference duration, the identification server determines that the response audio data packet meets the preset automatic response parameter.

Specifically, the preset automatic response parameter is a preset continuous voice reference duration, and the preset continuous voice reference duration may be set to 3.0s or 4.0s, or may be obtained by any other user through actual detection. For example, the identifier of the preset continuous voice reference duration is 3.0s, and the identification server analyzes the received response audio data packet sent by the answering device in the calling system, and obtains the received response audio data packet from the received response audio data packet. The longest continuous speech duration, if the longest continuous speech duration is 3.6 s, the longest continuous speech duration obtained is 3.6 s greater than the preset continuous speech reference duration of 3.0 s, and the recognition server determines the longest continuous The response audio data packet with a voice duration of 3.6 s satisfies the preset automatic response parameter, that is, the recognition server determines that the voice response type corresponding to the longest continuous voice duration of 3.6 s is an automatic response; if the longest acquisition is obtained The continuous speech duration is 1.6s, and the continuous speech duration is 1.6s less than the continuous speech reference duration of 3.0s. The device determines that the response audio packet with the longest continuous speech duration of 1.6s does not satisfy the preset automatic response parameter, that is, the recognition server determines that the voice response type corresponding to the longest continuous speech duration of 1.6s is artificial. Answer. In other embodiments of the present invention, the preset automatic response parameter may also be a continuous voice reference duration within a preset time period.

Obtaining, by analyzing the response audio data packet, a longest continuous voice duration from the received response audio data packet, where the recognition server determines the response audio when the acquired continuous voice duration is greater than a preset continuous voice reference duration The data packet satisfies the preset automatic response parameter, and determines that the voice response type corresponding to the response audio data packet is an automatic response. The realization recognizes the automatic response in the called response, so that the control device in the calling system controls the business process of the calling system according to the recognition result.

As shown in FIG. 5, it is a specific architectural diagram of a first embodiment of an identification device for automatically answering a call system of the present invention. The device is disposed in the identification server, and includes: the data transmission module 10 and the identification Module 20, wherein

The data transmission module 10 generally refers to a communication interface of the identification server, configured to establish an audio communication channel with the response device when receiving a response identification request sent by the answering device in the call system;

Receiving a response audio data packet sent by the response device based on the established audio communication channel.

Specifically, the call flow of the call system is started, the answering device in the call system receives the media from the phone user, and sends the called response audio data packet based on the received media, and the answering device sends the response identification to the data transmitting module 10 The request, when receiving the response identification request, the data transmitting module 10 establishes an audio communication channel with the answering device. The response device includes at least one communication channel for receiving media from a telephone user; the communication channel for receiving media from the telephone user corresponds to a communication channel provided with a response audio data packet; and the data transmission module 10 respectively Establishing an audio communication channel with the response device; for example, when the response device receives the media of the telephone user through the A communication channel, the response device transmits the called response audio data packet based on the received media, The communication channel sent by the response audio data packet is a B communication channel set with the A communication channel mapping; the response device sends a response identification request to the data transmission module 10, and when receiving the response identification request, the data transmission module 10 Establishing an audio communication channel with the B communication channel of the answering device; as described above, the data transmitting module 10 responds to the number of communication channels of the answering device in response to the audio data packet - correspondingly establishing audio with the answering device Communication channel.

The identification module 20, generally referred to as a processor of the identification server, is configured to analyze the response audio data packet to determine whether the response audio data packet satisfies a preset automatic response parameter;

When it is determined that the response audio data packet meets the preset automatic response parameter, it is determined that the voice response type corresponding to the response audio data packet is an automatic response. Specifically, based on the established audio communication channel with the response device, the data transmission module 10 receives the response audio data packet sent by the response device, and the identification module 20 analyzes the response audio data packet to determine Whether the response audio data packet meets the preset automatic response parameter, and when determining that the response audio data packet meets the preset automatic response parameter, the identification module 20 determines that the voice response type corresponding to the response audio data packet is an automatic response; When it is determined that the response audio data packet does not satisfy the preset automatic response parameter, the identification server determines that the voice response type corresponding to the response audio data packet is a manual response. In this embodiment, the preset automatic response parameter may be a preset mute reference ratio, or may be a preset continuous mute reference duration or a preset continuous speech reference duration, and the like. In the other embodiments of the present invention, the preset automatic response parameter may also be a mute reference ratio or a continuous mute reference duration or a preset time period. The continuous voice reference duration, etc., the preset time may be an applicable duration set by a user such as 30s or 40s in advance. For example, an audio communication channel C is established between the data transmitting module 10 and the answering communication channel B of the answering device, and the data transmitting module 10 receives the answering audio data from the answering communication channel B in the C audio communication channel receiving call system. In the package D, the preset automatic response parameter takes a preset mute reference ratio as an example, and the preset mute reference ratio is 25% as an example, and the identification module 20 analyzes the response audio data packet D to obtain the response audio data. The mute ratio in the packet D, if the obtained mute ratio in the response audio data packet D is 20%, the mute ratio 20% obtained by the recognition module 20 is less than 25% of the preset mute reference ratio, and the identification module 20 determines the response. The audio data packet D satisfies the preset automatic response parameter, that is, the identification module 20 determines that the voice response type corresponding to the response audio data packet of the silence ratio is 20% is an automatic response; if the acquired silence ratio in the response audio data packet D 30%, the acquired mute ratio is 30% greater than the preset mute reference ratio of 25%, and the identification module 20 determines that the response audio data packet D does not satisfy the preset automatic response parameter, ie, Module 20 determines the ratio of 30% mute audio packet corresponding to the response voice response to manual answer type. The data transmission module 10 sets the voice response type to automatic The identification result of the answer is sent to the control device in the call system, so that the control device controls the business process of the call system according to the recognition result.

The data transmitting module 10 receives the response audio data packet sent by the answering device through the established audio communication channel with the answering device in the calling system; the identifying module 20 analyzes the answering audio data packet to determine the answering audio data. Whether the packet satisfies the preset automatic response parameter, when determining that the response audio data packet satisfies the preset automatic response parameter, the identification module 20 determines that the voice response type corresponding to the response audio data packet is an automatic response. The realization recognizes the automatic response in the called response, so that the control device in the calling system controls the business process of the calling system according to the recognition result.

As shown in FIG. 6, it is a specific architecture diagram of a second embodiment of the identification device for automatically answering the call system of the present invention. The device includes a processing module 30,

The processing module 30 is configured to close the audio communication channel, perform deletion detection on the saved data, and delete data that meets the preset deletion condition.

Specifically, after determining that the response audio data packet meets the preset automatic response parameter, after the identification module 20 determines that the voice response type corresponding to the response audio data packet is an automatic response, the processing module 30 closes the response and the response. The audio communication channel established between the devices, and the deleted data is deleted and detected, and the data that meets the preset deletion condition is deleted. Taking the audio communication channel C established between the data transmission module 10 and the response communication channel B of the response device as an example, receiving the response audio data packet D sent by the response device through the audio communication channel C, After the identification module 20 determines that the response audio data packet D satisfies the automatic response parameter, the processing module 30 closes the audio channel channel C; the preset deletion condition may be a reference storage duration for saving data, or may be a user. The other parameters for saving the data set in advance, the preset deletion condition is an example of the storage duration of the saved data, and the processing module 30 detects the storage duration of the saved data, and if the saved data is found to be longer than the reference save time The processing module 30 deletes the found saved data, and the reference save duration may be 10 Days or 15 days, it can also be other reference save durations set by the user in advance. In another embodiment of the present invention, when the preset time is reached, the processing module 30 deletes and deletes the saved data, and deletes the data that meets the preset deletion condition, and is not limited to when a communication channel is closed. The processing module 30 performs the deletion detection on the saved data, and the preset time may be 1 day or 5 days, or may be any other time interval or time point set by the user in advance.

After receiving the response audio data packet sent by the answering device by the data transmitting module 10 and identifying by the identification module, the processing module 30 closes the established communication channel, and strives to lower the running load of the identifying device, thereby increasing the running speed, and By deleting the saved data of the identification device, the storage space of the identification device is utilized reasonably, and the processing speed is improved.

Preferably, the identification module 20 is further configured to obtain a continuous silent duration of the audio from the response audio data packet;

When the obtained continuous mute duration is less than the preset continuous mute reference duration, it is determined that the answer audio data packet satisfies the preset auto answer parameter.

Specifically, the preset automatic response parameter is a preset continuous mute reference duration, and the preset continuous mute reference duration may be set to 0.7s or 1.2s, or may be obtained by any other user through actual detection. For example, the identifier module 20 analyzes the received response audio data packet sent by the response device, and obtains the longest response voice packet from the received response audio data packet. The continuous mute duration, if the longest continuous mute duration is 0.6s, the longest continuous mute duration is 0.6s less than the preset continuous mute reference duration of 0.7s, and the recognition module 20 determines the longest continuous mute. The response audio data packet with a duration of 0.6 s satisfies the preset automatic response parameter, that is, the identification module 20 determines that the voice response type corresponding to the longest continuous silent silence duration of 0.6 s is an automatic response, if the longest acquisition is obtained. The continuous mute duration is 1.6s, and the obtained continuous mute duration is 1.6s longer than the continuous mute reference duration of 0.7s, and the identification module 20 determines the longest continuous mute duration. The response audio data packet for 1.6s does not satisfy the preset automatic response parameter, that is, the identification module 20 determines that the longest continuous silent duration is The response voice type of the 1.6s response audio packet is a manual response. In other embodiments of the present invention, the preset automatic response parameter may also be a preset continuous mute reference duration within a preset time period.

The recognition audio data packet is analyzed by the identification module 20, and the longest continuous silent duration is obtained from the received response audio data packet. When the acquired continuous silent duration is less than the preset continuous silent reference duration, the identification module 20 determines The response audio data packet satisfies a preset automatic response parameter, and determines that the voice response type corresponding to the response audio data packet is an automatic response. The automatic response in the called response is recognized so that the control device in the calling system controls the business process of the calling system according to the recognition result.

Preferably, the identification module 20 is further configured to acquire a continuous voice duration of audio from the response audio data packet; and

When the acquired continuous voice duration is greater than the preset continuous voice reference duration, it is determined that the response audio packet satisfies the preset automatic response parameter.

Specifically, the preset automatic response parameter is a preset continuous voice reference duration, and the preset continuous voice reference duration may be set to 3.0s or 4.0s, or may be obtained by any other user through actual detection. For example, the identification module 20 analyzes the received response audio data packet sent by the answering device in the call system, and receives the response audio data packet from the received response audio data packet. Obtaining the longest continuous speech duration, if the longest continuous speech duration is 3.6 s, the acquired continuous speech duration is 3.6 s greater than the preset continuous speech reference duration of 3.0 s, and the recognition module 20 determines the longest continuous speech duration. The response audio data packet with a duration of 3.6 s satisfies the preset automatic response parameter, that is, the recognition module 20 determines that the voice response type corresponding to the longest continuous voice duration of 3.6 s is an automatic response; if the longest acquisition is obtained The continuous speech duration is 1.6s, and the obtained continuous speech duration is 1.6s less than the continuous speech reference duration of 3.0s, and the recognition module 20 determines the longest continuous When sound when the audio data packet length of the response does not satisfy the preset 1.6s automatic answering parameters, i.e. the identification module 20 determines the longest continuous speech The voice response type corresponding to the 1.6S response audio packet is a manual response. In other embodiments of the present invention, the preset automatic response parameter may also be a preset continuous voice reference duration within a preset time period.

The recognition audio data packet is analyzed by the identification module 20, and the longest continuous speech duration is obtained from the received response audio data packet. When the acquired continuous speech duration is greater than the preset continuous speech reference duration, the identification module 20 determines The response audio data packet satisfies a preset automatic response parameter, and determines that the voice response type corresponding to the response audio data packet is an automatic response. The automatic response in the called response is recognized so that the control device in the calling system controls the business process of the calling system according to the recognition result.

The above description is only the preferred embodiment of the present invention, and is not intended to limit the scope of the invention, and the equivalent structure or equivalent flow transformation made by the specification and the drawings of the present invention may be directly or indirectly applied to other related The technical field is equally included in the scope of patent protection of the present invention.

Claims

claims

1. An identification method for automatic answering of a call system. The method includes:

When receiving a response identification request sent from the answering device in the calling system, the identification server establishes an audio communication channel with the answering device;

Based on the established audio communication channel, the identification server receives the response audio data packet sent by the response device;

The recognition server analyzes the response audio data packet to determine whether the response audio data packet satisfies the preset automatic response parameters. When it is determined that the response audio data packet satisfies the preset automatic response parameters, the recognition server determines the response. The voice response type corresponding to the audio data packet is automatic response.

2. The method for identifying automatic responses of a calling system according to claim 1, wherein when it is determined that the response audio data packet satisfies the preset automatic response parameters, the recognition server determines that the response audio data packet corresponds to After the step of setting the voice response type to automatic response, the method also includes:

The recognition server sends the recognition result that the voice response type is automatic response to the control device in the calling system, so that the control device controls the business process of the calling system based on the recognition result.

3. The method for identifying automatic responses of a calling system according to claim 1, wherein when it is determined that the response audio data packet satisfies the preset automatic response parameters, the recognition server determines that the response audio data packet corresponds to After the step of setting the voice response type to automatic response, the method also includes:

The recognition server closes the audio communication channel, performs deletion detection on the saved data, and deletes data that meets the preset deletion conditions.

4. The method for identifying automatic responses of a calling system according to claim 1 or 2, wherein the preset automatic response parameter is a preset continuous mute reference duration, and the identification server divides The step of analyzing the response audio data packet to determine whether the response audio data packet meets the automatic response parameters includes:

The identification server obtains the continuous audio silence duration from the response audio data packet;

When the obtained continuous silence duration is less than the preset continuous silence reference duration, the recognition server determines that the response audio data packet meets the preset automatic response parameters.

5. The method for identifying automatic responses of a calling system according to claim 1 or 2, wherein when the preset automatic response parameter is a preset continuous voice reference duration, the identification server analyzes the response audio data packet. , the steps to determine whether the response audio data packet meets the automatic response parameters include:

The recognition server obtains the continuous voice duration of the audio from the response audio data packet;

When the obtained continuous voice duration is greater than the preset continuous voice reference duration, the recognition server determines that the response audio data packet meets the preset automatic response parameters.

6. A device for identifying the automatic response of a call system. The device includes:

The data receiving and receiving module is configured to establish an audio communication channel with the answering device when receiving a response identification request sent by the answering device in the calling system; and

Based on the established audio communication channel, receive the response audio data packet sent by the response device;

The identification module is configured to analyze the response audio data packet to determine whether the response audio data packet satisfies the preset automatic response parameters; when it is determined that the response audio data packet satisfies the preset automatic response parameters, determine the The voice response type corresponding to the response audio data packet is automatic response.

7. The device for identifying automatic responses in a calling system according to claim 6, wherein the data receiving and receiving module is further configured to send the recognition result that the voice response type is automatic response to the control device in the calling system, so that all The control device controls the business process of the calling system according to the identification result.

8. The identification device for automatic answering of a call system according to claim 6, wherein the device further includes a processing module,

The processing module is configured to close the audio communication channel, perform deletion detection on the saved data, and delete data that meets preset deletion conditions.

9. The identification device for automatic answering of a calling system according to claim 6 or 7, wherein the preset automatic answering parameter is a preset continuous mute reference duration,

The identification module is also configured to obtain the continuous mute duration of the audio from the response audio data packet; and

When the obtained continuous silence duration is less than the preset continuous silence reference duration, it is determined that the response audio data packet meets the preset automatic response parameters.

10. The identification device for automatic answering of a calling system according to claim 6 or 7, wherein when the preset automatic answering parameter is a preset continuous voice reference duration,

The identification module is also configured to obtain the continuous voice duration of the audio from the response audio data packet; and

When the obtained continuous voice duration is greater than the preset continuous voice reference duration, it is determined that the response audio data packet meets the preset automatic response parameters.