CN111416911A - Calling platform of outbound system based on artificial intelligence - Google Patents

Abstract

The invention discloses an outbound call system call platform based on artificial intelligence. In order to overcome the problem that the outgoing call system in the prior art cannot identify the outgoing object and the outgoing content is easily leaked; the present invention adopts a central processing module and a verification voice sending module connected with the central processing module, and sends out the verification voice before the outgoing voice is played. ;Verification voice receiving module, which receives the verification voice reply of the dialed object; Personal voice database storage module, which stores the voiceprint information of the dialed object; Verification and comparison module, which compares the voiceprint to verify the identity of the dialed object; The verification result of the verification and comparison module is sent to the central processing module. The system can not only perform outgoing voice calls, but also identify whether the recipient is a person who needs to dial out through voice verification, so as to avoid the leakage of outgoing dialing content when necessary, and improve the security of the system.

Description

A call platform for outbound call system based on artificial intelligence

technical field

The invention relates to the technical field of calling, in particular to an outbound call system call platform based on artificial intelligence.

Background technique

The outbound call system refers to dialing the user's phone through the computer, and playing the recorded voice to the user through the computer. It is an integral part of modern customer service center system based on CTI technology. The outbound call is divided into two stages: the acquisition of the outbound call data and the initiation of the outbound call action. However, the existing outbound call system cannot identify the outbound object, and the outbound content is easily leaked, so it cannot be used in some special scenarios.

For example, a "voice outbound call system" disclosed in Chinese patent documents, its bulletin number "CN106572087B", includes a middleware server cluster, a set of databases and at least one outbound call server; wherein the middleware server cluster obtains external The outbound call task is recorded in the database, and the outbound call server retrieves the outbound call task from the database and initiates the outbound call. Although the system realizes the automation of outgoing calls, it cannot identify the object of outgoing calls, and the content of outgoing calls is easily leaked, resulting in poor security.

SUMMARY OF THE INVENTION

The invention mainly solves the problem that the outbound call system in the prior art cannot identify the outbound object, and the outbound content is easily leaked; it provides an artificial intelligence-based outbound system call platform, which recognizes the outbound object through voice verification, and ensures that the outbound content does not leak. It is leaked to outsiders, which improves the security of the system.

The above-mentioned technical problems of the present invention are mainly solved by the following technical solutions:

The present invention includes a central processing module, an outgoing voice sending module and a voice recognition module; the outgoing voice sending module and the voice recognition module are both connected with the central processing module; the system also includes

The verification voice sending module is connected with the central processing module, and the verification voice is sent to verify the identity of the outgoing object before the outgoing voice is played;

The verification voice receiving module is connected with the central processing module, and receives the verification voice reply of the dialed object after issuing the verification voice;

The personal voice database storage module, connected with the central processing module, stores the voiceprint information of the outgoing object;

The verification and comparison module is connected with the verification voice receiving module and the personal voice data storage module to verify the identity of the outgoing object;

The information feedback module is connected with the verification and comparison module and the central processing module, and sends the verification result of the verification and comparison module to the central processing module.

The system sends out the verification voice through the verification voice sending module before playing the outgoing voice, and the verification voice receiving module obtains the verification voice reply of the dialed object and sends it to the verification comparison module. The voice pattern of the object preset in the personal voice data storage module is compared to determine whether the answering object is the object required for outbound dialing, and the judgment result is sent to the central processing module through the information feedback module, and the central processing module is based on the judgment result. Controls whether the outgoing voice sending module plays the outgoing voice. Identify the identity of the recipient to ensure that the outgoing content is not leaked to outsiders, the system is simple and reliable, and the security of the system is improved.

Preferably, the system further includes a voice acquisition module and a voice preprocessing module; the voice acquisition module is connected to the voice preprocessing module, and the voice preprocessing module is connected to the voice recognition module; the voice preprocessing module acquires the voice acquisition module The voice information is de-cluttered, and the de-cluttered voice information is passed to the speech recognition module. The voice preprocessing module removes impurities from the received voice, which facilitates the subsequent voice recognition module to identify the voice content of the dialed object and improves the accuracy of the recognition.

Preferably, the voice recognition module is used to identify the voice information after removal of impurities, and identify the content of the dialed voice; and identify the identity information of the dialed voice object according to the dialed number, and transmit the identification information to the central processing unit. module. The speech recognition module recognizes the content of the outgoing voice and the identity of the outgoing target, which provides a prerequisite for the subsequent authentication of the outgoing target.

Preferably, the system further comprises an outgoing voice data storage module and a verification information storage module; the outgoing voice data storage module and the verification information storage module are both connected to the central processing module; the outgoing voice data storage module is used for The identification information transmitted to the central processing module is stored; the verification information storage module is used to store the identification information of the outgoing object and the verification result. The content stored in the outbound voice data storage module and the verification information storage module is convenient for subsequent data query, and also provides direction and data support for subsequent system upgrades.

Preferably, the verification voice sending module sends out verification voices that need to be repeated by the dialed object. The outgoing object can verify the identity by repeating the content of the verification voice as the verification voice reply, and through the comparison of the voiceprint. Sound verification to improve system security.

Preferably, the verification voice sending module sends out a preset verification voice that needs to be answered by the dialed object. The outgoing caller can verify the voice response by answering the content of the verification voice. The identity verification is realized by comparing the answer content and the voiceprint. The identity verification process is rigorous, ensuring the identity of the recipient and improving the security of the system.

Preferably, the verification and comparison module compares the verification voice reply of the outbound object obtained by the verification voice receiving module with the voiceprint of the outbound object in the personal voice data storage module to determine whether the identity of the outbound object is correct, and the identity verification result passes The information feedback module feeds back to the central processing module.

The central processing module controls whether the outgoing voice sending module plays the outgoing voice according to the judgment of the identity of the outgoing dialing object, so as to ensure that the content of the outgoing voice is not leaked to outsiders, and to ensure the security and privacy of the system.

The beneficial effects of the present invention are: through the module setting of the outbound call system, the system can not only make outbound voice calls, but also can identify whether the receiver is a person who needs to dial out through voice verification, so that the content of outbound calls can be avoided when needed. leaks, improving the security of the system.

Description of drawings

FIG. 1 is a schematic diagram of connection of a call platform of an outbound call system according to the present invention.

Figure 1. Voice acquisition module, 2. Voice preprocessing module, 3. Voice recognition module, 4. Central processing module, 5. Outgoing voice sending module, 6. Verification voice sending module, 7. Verification voice receiving module, 8 . Verification and comparison module, 9. Information feedback module, 10. Outgoing voice data storage module, 11. Verification information storage module, 12. Personal voice data storage module.

Detailed ways

The technical solutions of the present invention will be further described in detail below through embodiments and in conjunction with the accompanying drawings.

Example 1:

An artificial intelligence-based outbound call system call platform in this embodiment, as shown in FIG. 1 , includes a voice acquisition module 1, a voice preprocessing module 2, a voice recognition module 3, a central processing module 4, and an outbound voice sending module 5 , Verification voice sending module 6, verification voice receiving module 7, verification comparison module 8, information feedback module 9, outgoing voice data storage module 10, verification information storage module 11 and personal voice data storage module 12.

The speech acquisition module 1 is connected with the speech preprocessing module 2 , and the speech preprocessing module 2 is connected with the speech recognition module 3 . The voice recognition module 3, the dialing voice sending module 5, the verification voice sending module 6, the verification voice receiving module 7, the information feedback module 9, the dialling voice data storage module 10, the verification information storage module 11 and the personal voice data storage module 12 are all Connected to the central processing module 4 . The verification voice receiving module 7 is connected with the verification and comparison module 8 , the personal voice data storage module 12 is connected with the verification and comparison module 8 , and the verification and comparison module 8 is connected with the information feedback module 9 .

The speech preprocessing module 2 performs decontamination processing on the speech information acquired by the speech acquisition module 1 , and transmits the decontaminated speech information to the speech recognition module 3 . The speech preprocessing module 2 removes impurities from the received speech, which is convenient for the speech recognition module 3 to recognize the speech content of the dialed object and improves the recognition accuracy.

The voice recognition module 3 is used to identify the voice information after removal of impurities, and identify the content of the dialed voice; The outgoing voice data storage module 10 is used for storing the identification information transmitted to the central processing module 1 . The voice recognition module 3 recognizes the content of the dialed voice and the identity of the dialed object, and provides a prerequisite for the subsequent identity verification of the dialed object.

The central processing module 4 controls the verification voice sending module 6 to send a verification voice to verify the identity of the dialed object before the outgoing voice is played; the verification voice sending module 6 sends a verification voice that needs to be repeated by the dialed object. The verification voice receiving module 7 receives the verification voice reply of the outgoing object after the verification voice sending module 6 sends out the verification voice.

The personal voice database storage module 12 stores the voiceprint information of the dialed object, and the verification and comparison module 8 verifies the identity of the dialed subject according to the comparison between the received voice of the verification voice reply and the voiceprint in the personal voice database 12. The information feedback module 9 sends the verification result of the verification to the ratio module 8 to the central processing module 4 . The verification information storage module 11 is used to store the identity information of the outgoing object and the verification result.

The content stored in the outbound voice data storage module and the verification information storage module is convenient for subsequent data query, and also provides direction and data support for subsequent system upgrades.

The central processing module 4 controls whether the external voice sending module 5 plays the outgoing voice according to the identity verification result fed back by the information feedback module 9 . If the authentication result is correct, the outgoing voice is played, and if the authentication is incorrect, the outgoing voice is not played.

The voiceprint data of the dialed object is stored in the personal voice data storage module 12 in advance, and when it is necessary to dial out, the content of the voice information to be dialed and the information of the dialed object are called into the system, and the voice content can be recognized by the voice recognition module 3 and outbound objects.

Afterwards, the central processing module 4 stores the voice content that needs to be dialed out in the dial-out voice data storage module 10, and controls the verification voice sending module 6 to send out the verification voice that needs to be repeated to the dialed object, when the dialed object repeats the verification voice content When, by the verification voice receiving module 7, the verification voice reply sent by the outgoing object is received, and the verification voice reply issued by the outgoing object is compared with the voiceprint of the corresponding person stored in the personal voice data storage module 12 by the verification comparison module 8. right. When it is judged to be the same person, it will be fed back to the central processing module 4, and the central processing module 4 controls the outgoing voice sending module 5 to send out the voice content stored in the outgoing voice data storage module 10; when it is judged that it is not the same person, that is, no The system can not only make outgoing voice calls, but also identify whether the receiver is a person who needs to dial out, so as to avoid the leakage of outgoing dialing content when necessary, making the use safe.

Embodiment 2:

The artificial intelligence-based outbound call system call platform of this embodiment differs only in the verification content of the verification voice.

The central processing module 4 stores the voice content that needs to be dialed out in the dial-out voice data storage module 10, and controls the verification voice sending module 6 to send out the verification voice that needs to be answered by the dial-out object, and when the dial-out object answers the verification voice content, The verification voice reply sent by the outgoing object is received by the verification voice receiving module 7, and the verification voice reply issued by the outgoing object is compared with the voice pattern and content of the corresponding person stored in the personal voice data storage module 12 through the verification comparison module 8. right. When it is judged to be the same person, it will be fed back to the central processing module 4, and the central processing module 4 controls the outgoing voice sending module 5 to send out the voice content stored in the outgoing voice data storage module 10; when it is judged that it is not the same person, that is, no The system can not only make outgoing voice calls, but also identify whether the receiver is a person who needs to dial out, so as to avoid the leakage of outgoing dialing content when necessary, making the use safe.

Through the module setting of the outgoing call system, the system can not only make outgoing voice calls, but also identify whether the receiver is a person who needs to dial out through voice verification, so as to avoid the leakage of outgoing dialing content when necessary, and improve the system performance. security.

Claims (8)

1. An outbound system calling platform based on artificial intelligence comprises a central processing module (4), an outbound voice sending module (5) and a voice recognition module (3); the dial-out voice sending module (5) and the voice recognition module (3) are both connected with the central processing module (4); characterized in that the system also comprises

The verification voice sending module (6) is connected with the central processing module (4) and sends verification voice to verify the identity of the outbound object before the outbound voice is played;

The verification voice receiving module (7) is connected with the central processing module (4) and receives a verification voice reply of the dial-out object after sending out the verification voice;

The personal voice database storage module (12) is connected with the central processing module (4) and stores the voice print information of the dialing object;

The verification comparison module (8) is connected with the verification voice receiving module (7) and the personal voice data storage module (12) and is used for verifying the identity of the outbound object;

And the information feedback module (9) is connected with the verification comparison module (8) and the central processing module (4) and sends the verification result of the verification comparison module (8) to the central processing module (4).

2. The call platform of the outbound system based on artificial intelligence of claim 1, wherein the system further comprises a voice acquisition module (1) and a voice preprocessing module (2); the voice acquisition module (1) is connected with the voice preprocessing module (2), and the voice preprocessing module (2) is connected with the voice recognition module (3); the voice preprocessing module (2) removes impurities from the voice information acquired by the voice acquisition module (1), and transmits the voice information after removing the impurities to the voice recognition module (3).

3. The calling platform of the outbound system based on artificial intelligence as claimed in claim 2, wherein the voice recognition module (3) is used for recognizing the voice information after impurity removal and recognizing outbound voice content; and identifying the identity information of the outbound voice object according to the outbound number, and transmitting the identification information to the central processing module (1).

4. An artificial intelligence based outbound system call platform according to claim 3, characterised in that said system further comprises an outbound voice data storage module (10) and a verification information storage module (11); the dial-out voice data storage module (10) and the verification information storage module (11) are both connected with the central processing module (4); the dial-out voice data storage module (10) is used for storing the identification information transmitted to the central processing module (1); the verification information storage module (11) is used for storing the identity information of the outbound object and the verification result.

5. The call platform of outbound system based on artificial intelligence as claimed in claim 1, wherein said verification voice emitting module (6) emits verification voice that needs to be repeated by the outbound object.

6. The call platform of the outbound system based on artificial intelligence as claimed in claim 1, wherein said verification speech sound emitting module (6) emits a preset verification speech sound to be answered by the outbound party.

7. The call platform of the outbound system based on artificial intelligence as claimed in claim 5 or 6, wherein the verification comparison module (8) compares the verification voice reply of the outbound object obtained by the verification voice receiving module (7) with the voiceprint of the outbound object in the personal voice data storage module (12), determines whether the identity of the outbound object is correct, and feeds back the identity verification result to the central processing module (4) through the information feedback module (9).

8. The call platform of the outbound system based on artificial intelligence according to claim 7, wherein the central processing module (4) controls whether the outbound voice is played by the external voice sending module (5) according to the authentication result fed back by the information feedback module (9); if the identity authentication result is correct, the outbound voice is played, and if the identity authentication result is incorrect, the outbound voice is not played.

Images