A voiceprint recognition method and system for purifying a network environment during live broadcast

Abstract

The application provides a voiceprint recognition method for purifying a network environment during live broadcast, comprising: creating a comparison voiceprint library, including a white list voiceprint library, a key personnel voiceprint library and a speech and action problem voiceprint library; creating a live room voiceprint library, obtaining the voiceprint of each live room host in a live broadcast platform, and establishing the live room voiceprint library; registering the comparison voiceprint library; comparing the voiceprints, collecting the voiceprints of the live room hosts, and comparing the white list voiceprint library, the key personnel voiceprint library and the speech and action problem voiceprint library respectively; if there is a match with the key personnel voiceprint library, the corresponding host voiceprint collection and identification frequency is increased; if there is a match with the speech and action problem voiceprint library, the live room and the host are marked with a prohibited label, prohibited monitoring is performed, and automatic broadcast prohibition is performed when the prohibited number of times is reached; the method provided by the application binds the voiceprint with the host, creates, registers and compares various voiceprint libraries, combines the voiceprint recognition technology to purify the network environment, and reduces the borderline ball live broadcast.

Description

Technical Field

[0001] This invention relates to the field of voiceprint recognition, and in particular to a voiceprint recognition method for purifying the network environment during live streaming. Background Technology

[0002] With the rapid development of entertainment, live streaming and interactive methods are becoming increasingly diverse, attracting more and more young people to the field. Simultaneously, due to this youthful demographic, cybersecurity has become paramount, making it imperative to combat and prohibit streamers from skirting the rules. Currently, major live streaming platforms rely on manual review, random checks, or audience reports to curb such behavior. Some platforms have incorporated voice recognition and semantic analysis to combat this activity. However, these methods typically target a single live streaming account; while current technology may identify problematic streamers, the only solution is to ban their accounts.

[0003] The problem with banning an account is that it's possible to create another account, or even open a new account on another platform, to continue engaging in borderline live streams. Currently, there is no way to identify and resolve this issue. Summary of the Invention

[0004] The main objective of this invention is to overcome the aforementioned deficiencies in the prior art and propose a voiceprint recognition method for purifying the network environment during live streaming. This method binds voiceprints to the broadcaster and combines voiceprint recognition technology to purify the network environment and reduce borderline live streaming.

[0005] The present invention adopts the following technical solution:

[0006] A voiceprint recognition method for purifying the online environment during live streaming includes:

[0007] Create a voiceprint database for comparison, including a whitelist voiceprint database, a voiceprint database of key personnel, and a voiceprint database of speech and behavior issues;

[0008] Create a voiceprint database for live streaming rooms, obtain the voiceprints of each live streamer on the live streaming platform, and establish a voiceprint database for live streaming rooms.

[0009] The system registers and compares voiceprints with the corresponding live streamer's voiceprint database. First, it compares the voiceprint with the live streamer's voiceprint database. If the voiceprint matches the corresponding live streamer's voiceprint database, it then compares it with the voiceprint database for key personnel and the voiceprint database for speech and behavior issues. If there is no match, the voiceprint is registered in the whitelist voiceprint database. If the voiceprint does not match the corresponding live streamer's voiceprint database, it then compares it with the voiceprint database for key personnel and the voiceprint database for speech and behavior issues. If there is no match, the voiceprint is registered in the voiceprint database for key personnel.

[0010] Voiceprints are compared with those of the live streamers in the live stream. The voiceprints are compared with the whitelist voiceprint database, the key personnel voiceprint database, and the speech and behavior problem voiceprint database. If a match is found with the key personnel voiceprint database, the frequency of voiceprint collection and recognition for the corresponding streamer is increased. If a match is found with the speech and behavior problem voiceprint database, a prohibited label is added to the live stream and the streamer, and prohibited activities are monitored. If the number of prohibited activities reaches a certain number, the live stream is automatically banned.

[0011] Specifically, the aforementioned monitoring of prohibited activities includes:

[0012] Collect audio data from live streams that have been flagged as prohibited.

[0013] The voice data of live streams labeled with prohibited tags is identified using a speech recognition algorithm and compared with a prohibited word and phrase database. Specifically, before the prohibited monitoring step, the method also includes establishing a prohibited word and phrase database.

[0014] This invention provides a voiceprint recognition system for purifying the network environment during live streaming, comprising:

[0015] Voiceprint database creation unit: Creates voiceprint databases for comparison, including whitelist voiceprint databases, key personnel voiceprint databases, and speech and behavior issue voiceprint databases;

[0016] Live Streaming Room Voiceprint Database Creation Unit: Create a live streaming room voiceprint database, obtain the voiceprint of each live streamer in the live streaming platform, and establish the live streaming room voiceprint database;

[0017] Voiceprint database registration unit: Register and compare voiceprints to obtain the voiceprints of the live streamers. First, compare with the corresponding live streamer's voiceprint database. If it matches the corresponding live streamer's voiceprint database, then compare with the key personnel voiceprint database and the speech and behavior problem voiceprint database. If there is no match, register it in the whitelist voiceprint database. If it does not match the corresponding live streamer's voiceprint database, then compare with the key personnel voiceprint database and the speech and behavior problem voiceprint database. If there is no match, register it in the key personnel voiceprint database.

[0018] Voiceprint comparison and processing unit: compares voiceprints, collects the voiceprints of the live streamers, and compares them with the whitelist voiceprint database, the key personnel voiceprint database, and the speech and behavior problem voiceprint database. If a match is found in the key personnel voiceprint database, the frequency of voiceprint collection and recognition for the corresponding streamer is increased; if a match is found in the speech and behavior problem voiceprint database, a prohibited label is added to the live stream and the streamer, and prohibited activities are monitored. If the number of prohibited activities reaches a certain number, the live stream is automatically banned.

[0019] Specifically, the aforementioned monitoring of prohibited activities includes:

[0020] Collect audio data from live streams that have been flagged as prohibited.

[0021] The voice data of live streams labeled with prohibited tags is identified using a speech recognition algorithm and compared with a prohibited word and phrase database. Specifically, before the prohibited monitoring step, the method also includes establishing a prohibited word and phrase database.

[0022] In another aspect, the present invention provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-mentioned voiceprint recognition method for purifying the network environment in a live streaming room.

[0023] In another aspect, this invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the aforementioned voiceprint recognition method for purifying the network environment in a live streaming room.

[0024] As can be seen from the above description of the present invention, compared with the prior art, the present invention has the following beneficial effects:

[0025] This invention provides a voiceprint recognition method for purifying the network environment during live streaming, comprising: creating a comparison voiceprint database, including a whitelist voiceprint database, a key personnel voiceprint database, and a speech and behavior problem voiceprint database; creating a live streaming room voiceprint database by acquiring the voiceprint of each live streaming host on the live streaming platform; registering the comparison voiceprint database, acquiring the voiceprint of the live streaming host, first comparing it with the live streaming room voiceprint database, if it matches the corresponding live streaming room voiceprint database, then comparing it with the key personnel voiceprint database and the speech and behavior problem voiceprint database, if there is no match, registering it in the whitelist voiceprint database; if it does not match the corresponding live streaming room voiceprint database, then comparing it with the key personnel voiceprint database and the speech and behavior problem voiceprint database. The method involves comparing voiceprints with the voiceprint database. If no match is found, the voiceprint is registered in the key personnel voiceprint database. Voiceprints are collected from the livestream host and compared with the whitelist voiceprint database, the key personnel voiceprint database, and the speech and behavior problem voiceprint database. If a match is found with the key personnel voiceprint database, the frequency of voiceprint collection and recognition for the corresponding host is increased. If a match is found with the speech and behavior problem voiceprint database, a prohibited label is added to the livestream room and host, and prohibited activities are monitored. Once the number of prohibited activities reaches a certain threshold, the livestream is automatically banned. This invention binds voiceprints to hosts and, through the creation, registration, and comparison of multiple voiceprint databases, combines voiceprint recognition technology to purify the network environment and reduce borderline livestreaming. Attached Figure Description

[0026] Figure 1 A flowchart of a voiceprint recognition method for purifying the network environment during live streaming, provided by an embodiment of the present invention;

[0027] Figure 2 A flowchart illustrating the implementation of a voiceprint recognition method for purifying the network environment during live streaming, provided by an embodiment of the present invention;

[0028] Figure 3 This invention provides an architecture diagram of a voiceprint recognition system for purifying the network environment during live streaming, as provided in an embodiment of the present invention.

[0029] Figure 4A schematic diagram of an electronic device provided in an embodiment of the present invention;

[0030] Figure 5 This is a schematic diagram of an embodiment of a computer-readable storage medium provided in this invention.

[0031] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. Detailed Implementation

[0032] This invention provides a voiceprint recognition method for purifying the network environment during live streaming. It binds voiceprints to the broadcaster and purifies the network environment by creating, registering, and comparing multiple voiceprint databases, combined with voiceprint recognition technology, thereby reducing live streaming that skirts the edge of legality.

[0033] like Figure 1 This is a flowchart of a voiceprint recognition method for purifying the network environment during live streaming, according to an embodiment of the present invention. Figure 2 A flowchart illustrating the implementation of a voiceprint recognition method for purifying the network environment during live streaming, provided in this embodiment of the invention, specifically includes:

[0034] S101: Create a voiceprint database for comparison, including a whitelist voiceprint database, a voiceprint database of key personnel, and a voiceprint database of speech and behavior issues;

[0035] Due to the large number of live streaming platforms and designers, voiceprint databases are often created in multiple ways. These typically include a whitelist database, primarily used for registering the voiceprint data of ordinary streamers or users participating in voice interactions; a key personnel database, which mainly stores the voiceprints of streamers who have previously been involved in borderline content live streaming; and a speech and behavior problem database, which primarily stores the voiceprints of individuals whose speech has a negative impact.

[0036] S102: Create a voiceprint database for live streaming rooms, obtain the voiceprint of each live streamer on the live streaming platform, and establish a voiceprint database for live streaming rooms.

[0037] Currently, there are a huge number of live streaming rooms, and each live streaming room is equivalent to a small voiceprint database. First, a corresponding voiceprint database will be created for each live streaming room.

[0038] S103: Register and compare voiceprints to the voiceprint database to obtain the voiceprint of the live streamer. First, compare the voiceprint with the corresponding live stream voiceprint database. If it matches the corresponding live stream voiceprint database, then compare it with the voiceprint database of key personnel and the voiceprint database of speech and behavior issues. If there is no match, register it in the whitelist voiceprint database. If it does not match the corresponding live stream voiceprint database, then compare it with the voiceprint database of key personnel and the voiceprint database of speech and behavior issues. If there is no match, register it in the voiceprint database of key personnel.

[0039] After establishing the voiceprint database for comparison and the voiceprint database for the live stream, it is necessary to register the voiceprint database for comparison, registering the voiceprint database for the live stream into the voiceprint database for comparison, thus linking the two databases. For broadcasters whose voiceprints do not match those in the live stream database, their voiceprints should be registered in the key personnel voiceprint database for close monitoring.

[0040] S104: Voiceprint comparison. Collect the voiceprint of the live streamer and compare it with the whitelist voiceprint database, the key personnel voiceprint database, and the speech and behavior problem voiceprint database. If a match is found in the key personnel voiceprint database, the frequency of voiceprint collection and recognition for the corresponding streamer will be increased. If a match is found in the speech and behavior problem voiceprint database, a prohibited label will be added to the live stream and the streamer, and prohibited activities will be monitored. If the number of prohibited activities reaches a certain number, the live stream will be automatically banned.

[0041] Specifically, the aforementioned monitoring of prohibited activities includes:

[0042] Collect audio data from live streams that have been flagged as prohibited.

[0043] The speech recognition algorithm is used to identify the audio data of live streams that have been tagged with prohibited words and phrases, and the data is compared with a prohibited word and phrase database.

[0044] Speech recognition algorithms include, but are not limited to, algorithms based on Dynamic Time Warping, methods based on Hidden Markov Models (HMMs) with parametric models, and algorithms and hybrid algorithms based on neural networks (ANNs). Examples include convolutional neural networks, deep learning neural networks, backpropagation neural networks, RBF neural networks, and fuzzy clustering neural networks.

[0045] Specifically, prior to the prohibited monitoring step, the method also includes establishing a database of prohibited words and phrases.

[0046] The prohibited words and phrases database contains prohibited words and phrases and is updated regularly.

[0047] like Figure 3 This invention provides an architecture diagram of a voiceprint recognition system for purifying the network environment during live streaming, as provided in an embodiment of the present invention.

[0048] Voiceprint database creation unit 301: Creates a voiceprint database for comparison, including a whitelist voiceprint database, a key personnel voiceprint database, and a speech and behavior issue voiceprint database;

[0049] Due to the large number of live streaming platforms and designers, voiceprint databases are often created in multiple ways. These typically include a whitelist database, primarily used for registering the voiceprint data of ordinary streamers or users participating in voice interactions; a key personnel database, which mainly stores the voiceprints of streamers who have previously been involved in borderline content live streaming; and a speech and behavior problem database, which primarily stores the voiceprints of individuals whose speech has a negative impact.

[0050] Live Streaming Room Voiceprint Database Creation Unit 302: Create a live streaming room voiceprint database, obtain the voiceprint of each live streamer in the live streaming platform, and establish the live streaming room voiceprint database;

[0051] Currently, there are a huge number of live streaming rooms, and each live streaming room is equivalent to a small voiceprint database. First, a corresponding voiceprint database will be created for each live streaming room.

[0052] Voiceprint database registration unit 303: Register and compare voiceprint databases to obtain the voiceprint of the live streamer. First, compare with the corresponding live stream voiceprint database. If it matches the corresponding live stream voiceprint database, then compare with the key personnel voiceprint database and the speech and behavior problem voiceprint database. If there is no match, register it in the whitelist voiceprint database. If it does not match the corresponding live stream voiceprint database, then compare with the key personnel voiceprint database and the speech and behavior problem voiceprint database. If there is no match, register it in the key personnel voiceprint database.

[0053] After establishing the voiceprint database for comparison and the voiceprint database for the live stream, it is necessary to register the voiceprint database for comparison, registering the voiceprint database for the live stream into the voiceprint database for comparison, thus linking the two databases. For broadcasters whose voiceprints do not match those in the live stream database, their voiceprints should be registered in the key personnel voiceprint database for close monitoring.

[0054] Voiceprint comparison and processing unit 304: compares voiceprints, collects the voiceprints of the live broadcast room anchors, and compares them with the whitelist voiceprint database, the key personnel voiceprint database, and the speech and behavior problem voiceprint database. If a match is found in the key personnel voiceprint database, the frequency of voiceprint collection and recognition for the corresponding anchor is increased; if a match is found in the speech and behavior problem voiceprint database, a prohibited label is added to the live broadcast room and anchor, and prohibited monitoring is carried out. If the number of prohibited times is reached, the broadcast is automatically banned.

[0055] Specifically, the aforementioned monitoring of prohibited activities includes:

[0056] Collect audio data from live streams that have been flagged as prohibited.

[0057] The speech recognition algorithm is used to identify the audio data of live streams that have been tagged with prohibited words and phrases, and the data is compared with a prohibited word and phrase database.

[0058] Speech recognition algorithms include, but are not limited to, algorithms based on Dynamic Time Warping, methods based on Hidden Markov Models (HMMs) with parametric models, and algorithms and hybrid algorithms based on neural networks (ANNs). Examples include convolutional neural networks, deep learning neural networks, backpropagation neural networks, RBF neural networks, and fuzzy clustering neural networks.

[0059] Specifically, prior to the prohibited monitoring step, the method also includes establishing a database of prohibited words and phrases.

[0060] The prohibited words and phrases database contains prohibited words and phrases and is updated regularly.

[0061] Figure 4 As shown, this embodiment of the invention provides an electronic device 400, including a memory 410, a processor 420, and a computer program 411 stored in the memory 420 and executable on the processor 420. When the processor 420 executes the computer program 411, it implements a voiceprint recognition method for purifying the network environment in a live streaming room, as provided in this embodiment of the invention.

[0062] Since the electronic device described in this embodiment is the device used to implement the embodiments of the present invention, those skilled in the art can understand the specific implementation methods and various variations of the electronic device in this embodiment based on the methods described in the embodiments of the present invention. Therefore, how the electronic device implements the methods in the embodiments of the present invention will not be described in detail here. Any device used by those skilled in the art to implement the methods in the embodiments of the present invention is within the scope of protection of the present invention.

[0063] Please see Figure 5 , Figure 5 This is a schematic diagram of an embodiment of a computer-readable storage medium provided in this invention.

[0064] like Figure 5 As shown, this embodiment provides a computer-readable storage medium 500, on which a computer program 511 is stored. When the computer program 511 is executed by a processor, it implements a voiceprint recognition method for purifying the network environment in a live streaming room, as provided in this embodiment of the invention.

[0065] It should be noted that the descriptions of each embodiment in the above embodiments have different focuses. For parts that are not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0066] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0067] This invention provides a voiceprint recognition method for purifying the network environment during live streaming, comprising: creating a comparison voiceprint database, including a whitelist voiceprint database, a key personnel voiceprint database, and a speech and behavior problem voiceprint database; creating a live streaming room voiceprint database by acquiring the voiceprint of each live streaming host on the live streaming platform; registering the comparison voiceprint database, acquiring the voiceprint of the live streaming host, first comparing it with the live streaming room voiceprint database, if it matches the corresponding live streaming room voiceprint database, then comparing it with the key personnel voiceprint database and the speech and behavior problem voiceprint database, if there is no match, registering it in the whitelist voiceprint database; if it does not match the corresponding live streaming room voiceprint database, then comparing it with the key personnel voiceprint database and the speech and behavior problem voiceprint database. The method involves comparing voiceprints with the voiceprint database. If no match is found, the voiceprint is registered in the key personnel voiceprint database. Voiceprints are collected from the livestream host and compared with the whitelist voiceprint database, the key personnel voiceprint database, and the speech and behavior problem voiceprint database. If a match is found with the key personnel voiceprint database, the frequency of voiceprint collection and recognition for the corresponding host is increased. If a match is found with the speech and behavior problem voiceprint database, a prohibited label is added to the livestream room and host, and prohibited activities are monitored. Once the number of prohibited activities reaches a certain threshold, the livestream is automatically banned. This invention binds voiceprints to hosts and, through the creation, registration, and comparison of multiple voiceprint databases, combines voiceprint recognition technology to purify the network environment and reduce borderline livestreaming.

[0068] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element. The above descriptions are merely specific embodiments of this application, enabling those skilled in the art to understand or implement this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features claimed herein.

[0069] The above are merely specific embodiments of the present invention, but the design concept of the present invention is not limited thereto. Any non-substantial modifications made to the present invention using this concept shall be considered as infringing upon the protection scope of the present invention.

Claims

1. A voiceprint recognition method for purifying the network environment during live streaming, characterized in that, include: Create a voiceprint database for comparison, including a whitelist voiceprint database, a voiceprint database of key personnel, and a voiceprint database of speech and behavior issues; Create a voiceprint database for live streaming rooms, obtain the voiceprints of each live streamer on the live streaming platform, and establish a voiceprint database for live streaming rooms. The system registers and compares voiceprints with the corresponding live streamer's voiceprint database. First, it compares the voiceprint with the live streamer's voiceprint database. If the voiceprint matches the corresponding live streamer's voiceprint database, it then compares it with the voiceprint database for key personnel and the voiceprint database for speech and behavior issues. If there is no match, the voiceprint is registered in the whitelist voiceprint database. If the voiceprint does not match the corresponding live streamer's voiceprint database, it then compares it with the voiceprint database for key personnel and the voiceprint database for speech and behavior issues. If there is no match, the voiceprint is registered in the voiceprint database for key personnel. Voiceprints are compared with those of the live streamers in the live stream. The voiceprints are compared with the whitelist voiceprint database, the key personnel voiceprint database, and the speech and behavior problem voiceprint database. If a match is found with the key personnel voiceprint database, the frequency of voiceprint collection and recognition for the corresponding streamer is increased. If a match is found with the speech and behavior problem voiceprint database, a prohibited label is added to the live stream and the streamer, and prohibited activities are monitored. If the number of prohibited activities reaches a certain number, the live stream is automatically banned.

2. The voiceprint recognition method for purifying the network environment during live streaming according to claim 1, characterized in that, The prohibited content monitoring specifically involves: collecting audio data from live streams that have been tagged with prohibited content; using a speech recognition algorithm to identify the audio data from the live streams that have been tagged with prohibited content, and comparing it with a prohibited word and phrase database.

3. The voiceprint recognition method for purifying the network environment during live streaming according to claim 2, characterized in that, Before the prohibited monitoring step, the method also includes establishing a database of prohibited words and phrases.

4. A voiceprint recognition system for purifying the network environment during live streaming, characterized in that, include: Voiceprint database creation unit: Creates voiceprint databases for comparison, including whitelist voiceprint databases, key personnel voiceprint databases, and speech and behavior issue voiceprint databases; Live Streaming Room Voiceprint Database Creation Unit: Create a live streaming room voiceprint database, obtain the voiceprint of each live streamer in the live streaming platform, and establish the live streaming room voiceprint database; Voiceprint database registration unit: Register and compare voiceprint databases to obtain the voiceprint of the live streamer. First, compare with the corresponding live stream voiceprint database. If it matches the corresponding live stream voiceprint database, then compare with the key personnel voiceprint database and the speech and behavior problem voiceprint database. If there is no match, register it in the whitelist voiceprint database. If it does not match the corresponding live stream voiceprint database, then compare with the key personnel voiceprint database and the speech and behavior problem voiceprint database. If there is no match, register it in the key personnel voiceprint database. Voiceprint comparison and processing unit: compares voiceprints, collects the voiceprints of the live streamers, and compares them with the whitelist voiceprint database, the key personnel voiceprint database, and the speech and behavior problem voiceprint database. If a match is found in the key personnel voiceprint database, the frequency of voiceprint collection and recognition for the corresponding streamer is increased; if a match is found in the speech and behavior problem voiceprint database, a prohibited label is added to the live stream and the streamer, and prohibited activities are monitored. If the number of prohibited activities reaches a certain number, the live stream is automatically banned.

5. A voiceprint recognition system for purifying the network environment during live streaming, as described in claim 4, is characterized in that... The prohibited content monitoring specifically involves: collecting audio data from live streams that have been tagged with prohibited content; using a speech recognition algorithm to identify the audio data from the live streams that have been tagged with prohibited content, and comparing it with a prohibited word and phrase database.

6. A voiceprint recognition system for purifying the network environment during live streaming, as described in claim 5, is characterized in that... Before the prohibited monitoring step, the method also includes establishing a database of prohibited words and phrases.

7. An electronic device, characterized in that, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the steps of the method according to any one of claims 1-3.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the method described in any one of claims 1-3.

Images