Method and device for calculating aperture opening ratio based on VAD algorithm

Abstract

The invention discloses an aperture opening ratio calculation method and device based on a VAD algorithm. The method comprises the following steps: acquiring a sound recording file of an online classroom; based on a VAD algorithm, voice activity information of the teacher recording file and voice activity information of the student recording file are detected according to a preset time interval, and a first voice character string of the teacher recording file and a second voice character string of the student recording file are obtained; carrying out statistics on voice activity information in the first voice character string and the second voice character string, and according to a statistical result, carrying out calculation to obtain a student aperture opening ratio of the online classroom; the student aperture ratio is determined according to the ratio of the voice activity information in the second voice character string to the voice activity information in the first voice character string. The voice activities of teachers and students in the online class can be accurately and quickly identified in real time by using the VAD algorithm, so that the student aperture opening ratio in the online class can be accurately calculated, and the learning state of the students can be effectively monitored.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of speech recognition technology, and in particular to a method and device for calculating an aperture ratio based on a VAD algorithm. Background Art

[0002] Online education uses the Internet as a medium to realize online learning, eliminating the limitations of time and space, facilitating and flexible arrangements for learning, making it more convenient and flexible, and able to meet the needs of different students, thereby improving learning efficiency.

[0003] For online language learning, students not only need to learn the writing and rules of the language, but also need to practice using the language. The speaking rate of students reflects their acceptance, participation and mastery of language courses. It is an important learning indicator for online learning and is used to evaluate teaching.

[0004] When processing the opening rate, the existing technology can adopt a course recording method, such as performing ASR (Automatic Speech Recognition) analysis on the recording file after the course is completed, identifying the speech content of the teacher and the student at each time point (such as speech to text), and determining the opening rate based on the number of times the two speak. However, the cost of using ASR is high, requiring a large amount of computing resources, and the service fee is expensive. ASR analysis takes a long time, especially when there are recording files of multiple courses, a queuing mechanism needs to be used to identify them one by one, resulting in the inability to identify them in time after the course ends, affecting the timeliness and effectiveness of calculating the opening rate. Alternatively, real-time ASR processing is used, but a large number of courses will cause ASR concurrency problems, increase server load, and cause performance bottlenecks. Furthermore, real-time ASR requires a lot of resources, requires improving system stability and network bandwidth, and also increases the complexity and cost of technical implementation. Summary of the invention

[0005] In view of the above problems, embodiments of the present invention are proposed to provide a method and device for calculating aperture ratio based on a VAD algorithm that overcomes the above problems or at least partially solves the above problems.

[0006] According to one aspect of an embodiment of the present invention, a method for calculating an aperture ratio based on a VAD algorithm is provided, the method comprising:

[0007] Get the recording files of online classes; the recording files include the teacher's recording files and the students' recording files;

[0008] Based on the VAD algorithm, the voice activity information of the teacher's recording file and the student's recording file are detected at preset time intervals to obtain a first voice string of the teacher's recording file and a second voice string of the student's recording file; wherein the first voice string and the second voice string are composed of preset characters representing the voice activity information;

[0009] The voice activity information in the first voice string and the second voice string is counted, and the student opening rate in the online class is calculated based on the statistical results; the student opening rate is determined based on the ratio of the voice activity information in the second voice string to the first voice string.

[0010] According to another aspect of an embodiment of the present invention, there is provided an aperture ratio calculation device based on a VAD algorithm, comprising:

[0011] The acquisition module is suitable for acquiring the recording files of online classes; the recording files include the teacher's recording files and the student's recording files;

[0012] A speech recognition module, adapted to detect speech activity information of a teacher's recording file and a student's recording file respectively at preset time intervals based on a VAD algorithm, and obtain a first speech string of the teacher's recording file and a second speech string of the student's recording file; wherein the first speech string and the second speech string are composed of preset characters representing the speech activity information;

[0013] The calculation module is suitable for counting the voice activity information in the first voice string and the second voice string, and calculating the student opening rate in the online class according to the statistical results; the student opening rate is determined according to the ratio of the voice activity information in the second voice string to the first voice string.

[0014] According to another aspect of an embodiment of the present invention, there is provided a computing device, comprising: a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus;

[0015] The memory is used to store at least one executable instruction, and the executable instruction enables the processor to execute operations corresponding to the aperture ratio calculation method based on the VAD algorithm.

[0016] According to another aspect of the embodiments of the present invention, a computer storage medium is provided, wherein the storage medium stores at least one executable instruction, and the executable instruction enables a processor to perform operations corresponding to the aperture ratio calculation method based on the VAD algorithm as described above.

[0017] According to another aspect of the embodiments of the present invention, a computer program product is provided, comprising at least one executable instruction, wherein the executable instruction enables a processor to perform operations corresponding to the above-mentioned method for calculating aperture ratio based on the VAD algorithm.

[0018] According to the aperture rate calculation method and device based on the VAD algorithm provided in the embodiments of the present invention, the VAD algorithm can be used to accurately, real-time and quickly identify the voice activities of teachers and students in online classes, so that the aperture rate of students in online classes can be accurately calculated, thereby realizing effective monitoring of students' learning status.

[0019] The above description is only an overview of the technical solution of the embodiment of the present invention. In order to more clearly understand the technical means of the embodiment of the present invention, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the embodiment of the present invention more obvious and easy to understand, the specific implementation method of the embodiment of the present invention is specifically listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the embodiments of the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0021] Figure 1 A flow chart of a method for calculating aperture ratio based on a VAD algorithm according to an embodiment of the present invention is shown;

[0022] Figure 2 A schematic diagram of the structure of an aperture ratio calculation device based on a VAD algorithm according to an embodiment of the present invention is shown;

[0023] Figure 3 A schematic diagram of the structure of a computing device according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0024] The exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present invention and to enable the scope of the present invention to be fully communicated to those skilled in the art.

[0025] First, the terms involved in one or more embodiments of the present application are explained.

[0026] VAD: Voice Activity Detection algorithm is a technology used in speech processing to detect active speech and inactive speech parts in input audio signals. Its main purpose is to effectively distinguish useful speech signals from background noise in applications such as noise reduction, speech recognition, and speech encoding and decoding.

[0027] ASR: Automatic Speech Recognition technology, whose goal is to convert the vocabulary content in human speech into computer-readable input, such as keystrokes, binary codes, or character sequences.

[0028] Figure 1 FIG. 4 shows a flow chart of a method for calculating aperture ratio based on a VAD algorithm according to an embodiment of the present invention. Figure 1 As shown, the method comprises the following steps:

[0029] Step S101, obtaining the recording files of the online class; the recording files include the teacher's recording files and the student's recording files.

[0030] The opening rate of students in online classes can be used to measure the students' participation and then analyze the students' learning status. Timely and accurate calculation of the opening rate of students can also help online classes evaluate teaching.

[0031] When calculating the student opening rate in this embodiment, it is calculated by performing voice recognition on the recording files of the online class. The recording files of the online class include the teacher recording files and the student recording files. The teacher recording files and the student recording files record the process of online teaching. The teacher recording files only contain the teacher's voice, and the student recording files only contain the student's voice. The recording files can use different recording channels to record the voices of teachers and students respectively, so that when voice activity is recognized later, voice activity recognition can be performed directly for different recording files, without having to distinguish between teachers and students based on the voice. For example, different recording channels are constructed for the teacher client and the student client respectively. When the teacher speaks on the teacher client, the teacher's voice is recorded to obtain the teacher recording file. When the student speaks on the student client, the student's voice is recorded to obtain the student recording file, etc. The above is an example, which is set according to the implementation situation and is not limited here.

[0032] The teacher recording file and the student recording file can be stored on the server side. When the online class ends, the stored teacher recording file and the student recording file are directly obtained based on the server side for voice activity recognition. There is no need to distinguish whether the speaker in the recording file is a teacher or a student, and voice recognition can be performed more quickly. Further, this embodiment can quickly identify voice activity information and calculate the student's opening rate. When obtaining the teacher recording file and the student recording file, the teacher recording file and the student recording file of the entire online class can be used for voice activity recognition at the end of the online class to calculate the student's opening rate of the entire online class. It is also possible to obtain recording files at different stages of the online class. For example, the online class contains multiple classes or multiple teaching contents. When a class ends or a teaching content is completed, the teacher recording file and the student recording file of the class or teaching content are obtained, and the student's opening rate of the class or teaching content is calculated. It is convenient to review the opening rates of students at different stages, determine the learning situation of students at different stages of the online class, and then remind students to review online classes at different stages. The above is an example, which is specifically set according to the implementation situation and is not limited here.

[0033] Step S102: based on the VAD algorithm, the voice activity information of the teacher's recording file and the student's recording file are detected at preset time intervals to obtain a first voice string of the teacher's recording file and a second voice string of the student's recording file.

[0034] After obtaining the teacher's recording file and the student's recording file, the VAD algorithm can be used to detect the teacher's recording file and the student's recording file. During the detection, the VAD algorithm can efficiently distinguish between voice activities and non-voice activities without the need for complex voice recognition or recognition of semantics in the recording.

[0035] Considering that the recording files may contain various noises such as noise and background sound, before detecting the recording files, the voice activity information of the teacher's recording files and the student's recording files can be preprocessed. The voice preprocessing includes frequency division, DC offset removal, and pre-emphasis. Frequency division can divide the audio signal in the recording file into multiple short-term stable audio frames to simulate the short-term stability of the voice signal. The length of each frame can be set to 10ms-30ms, etc. Based on the audio frames obtained after frequency division, the DC offset can be removed from the audio frames to eliminate the DC component in the audio signal, improve the audio quality, reduce the background noise and distortion caused by the DC offset, and more accurately identify the voice activity. Pre-emphasis enhances the high-frequency components of the signal, compensates for the high-frequency attenuation during the transmission process, and improves the clarity and information content of the audio signal. The above voice preprocessing can facilitate more accurate identification of whether the audio frame contains voice activity during VAD algorithm detection.

[0036] Based on the VAD algorithm, feature extraction is performed on the teacher recording file and the student recording file after voice preprocessing, and the voice features of each audio frame are extracted. The extracted voice features include various voice features such as short-time energy, zero-crossing rate, spectral entropy, etc., and the voice activity information is determined based on the voice features. Specifically, for the teacher recording file after voice preprocessing, the VAD algorithm is used to extract features of each audio frame in the teacher recording file to obtain the voice features of each audio frame in the teacher recording file. Correspondingly, feature extraction is performed on each audio frame in the student recording file after voice preprocessing to obtain the voice features of each audio frame in the student recording file. The VAD algorithm can detect whether there is voice activity in the audio signal and distinguish between voice activity and non-voice activity. The VAD algorithm has low computing resource requirements, does not need to convert the entire voice signal into text, only needs to detect the voice activity therein, can more accurately identify voice activity, and has fast processing speed, low computing cost, no need to wait for a long time, and is suitable for a large number of concurrent online classroom recording file processing. Further, in addition to the VAD algorithm, other voice activity recognition algorithms can also be used to identify recording files, which are not limited here.

[0037] According to the extracted speech features of each audio frame, a cyclic judgment can be performed in sequence, such as cyclically judging whether the speech features of each audio frame in the teacher's recording file represent that it contains speech activity information. According to the judgment result, the preset characters representing the speech activity information are used to form a first speech string in the order of the audio frames. If it is judged that it contains speech activity information, the result corresponding to the audio frame is marked as the first character; if it does not contain speech activity information, the result corresponding to the audio frame is marked as the second character. Among them, the first character represents that the speech activity information is contained, and the second character represents that the language activity is not contained. Through cyclic judgment, until the judgment of the speech features of all audio frames in the teacher's recording file is completed, the first speech string of the teacher's recording file is obtained according to the order of the speech frames. Similarly, it is cyclically judged whether the speech features of each audio frame in the student's recording file represent that it contains speech activity information. According to the judgment result, the second speech string is formed by the preset characters representing the speech activity information in the order of the audio frames. To facilitate subsequent statistical calculations, the first character can be set to 1 and the second character to 0, so that subsequent statistics can be performed directly based on the values ​​of the first and second characters. It can also be set to a and b, etc. For subsequent statistics, the number of the first character can be counted. The above is an example, and the specific setting depends on the implementation situation and is not limited here.

[0038] For determining whether the speech feature of the audio frame indicates that it contains speech activity information, it can be determined by, for example, determining whether the speech feature of each audio frame is greater than a preset speech activity threshold. The preset speech activity threshold can be set according to the implementation situation, by comparing the values ​​contained in the speech feature of each audio frame with the preset speech activity threshold. If it is greater than the preset speech activity threshold, it is determined that the audio frame contains speech activity information; otherwise, it is determined that the audio frame does not contain speech activity information. If the speech activity information is contained, the result corresponding to the audio frame is marked as the first character; if the speech activity information is not contained, the result corresponding to the audio frame is marked as the second character. Alternatively, the speech features of each audio frame can be classified based on a preset classification algorithm to obtain audio frames containing speech activity information and audio frames not containing speech activity information. The preset classification algorithm can use a classifier such as a Gaussian mixture model, a neural network, etc. to construct a classifier, and the classifier is used to classify and determine the audio frames containing speech activity information and the audio frames not containing speech activity information. If the speech activity information is contained, the result corresponding to the audio frame is marked as the first character; if the speech activity information is not contained, the result corresponding to the audio frame is marked as the second character. The above is an example, and other classification algorithms can also be used, which will not be described in detail here.

[0039] Step S103, counting the voice activity information in the first voice string and the second voice string, and calculating the student speaking rate in the online class based on the statistical results.

[0040] After obtaining the first voice string and the second voice string, the voice activity information in the first voice string and the second voice string is counted to obtain the voice activity information of the teacher and the voice activity information of the students, and the student opening rate of the online class is calculated based on the voice activity information of the teacher and the voice activity information of the students. The student opening rate is determined based on the ratio of the voice activity information in the second voice string to the first voice string.

[0041] Specifically, in order to facilitate quick statistics, the first voice string and the second voice string can be divided according to the preset length to obtain multiple first voice substrings and multiple second voice substrings. For example, if the preset length is 10, the first voice string and the second voice string are divided in units of 10 to obtain multiple first voice substrings and multiple second voice substrings of length 10, and the number of substrings of the obtained first voice substring and the second voice substring is the same. If the online class time is 25 minutes, the length of the teacher's recording file and the student's recording file is also 25 minutes, the preset length is 10 seconds, the first voice string is divided to obtain 150 first voice substrings, and the second voice string is divided to obtain 150 second voice substrings.

[0042] Count the first data containing the first character in each first voice substring, and the second data containing the first character in each second voice substring. For example, when the first character is 1, the sum of 1s in each first voice substring can be directly counted to obtain the first data. If the first character is other characters, then count the number of first characters contained in each first voice substring to obtain the first data. Similarly, count the number of first characters contained in each second voice substring to obtain the second data. The first data represents the number of voice activities of the teacher, and the second data represents the number of voice activities of the students. The ratio of the second data to the sum is calculated based on the sum of the first data and the second data, and the average ratio is calculated based on the ratio and the number of substrings as the student opening rate in the online class. Refer to the following formula for details:

[0043]

[0044] Among them, n is the number of substrings, stu i is the number of the first character in the i-th second voice substring, tea i is the number of the first character in the i-th first voice substring, accumulating each stu i Get the second data, accumulate each tea i The first data is obtained. The ratio of the second data to the sum is calculated based on the sum of the first data and the second data, and then divided by the number of substrings. This can eliminate the influence of a single operation error, ensure the accuracy of the calculation result, and achieve effective management of the learning situation of students in the online class.

[0045] According to the aperture rate calculation method based on the VAD algorithm provided in an embodiment of the present invention, the VAD algorithm can be used to accurately, real-time and quickly identify the voice activities of teachers and students in online classes, so that the aperture rate of students in online classes can be accurately calculated, thereby achieving effective monitoring of students' learning status.

[0046] Figure 2 FIG. 2 shows a schematic diagram of the structure of an aperture ratio calculation device based on a VAD algorithm provided by an embodiment of the present invention. Figure 2 As shown, the device comprises:

[0047] The acquisition module 210 is adapted to acquire the recording files of the online class; the recording files include the teacher's recording files and the student's recording files;

[0048] The speech recognition module 220 is adapted to detect the speech activity information of the teacher's recording file and the student's recording file respectively at preset time intervals based on the VAD algorithm, and obtain a first speech string of the teacher's recording file and a second speech string of the student's recording file; wherein the first speech string and the second speech string are composed of preset characters representing the speech activity information;

[0049] The calculation module 230 is suitable for counting the voice activity information in the first voice string and the second voice string, and calculating the student opening rate in the online class based on the statistical results; the student opening rate is determined based on the ratio of the voice activity information in the second voice string to the first voice string.

[0050] Optionally, the preset characters include a first character representing that the voice activity information is included and a second character representing that the language activity is not included;

[0051] The speech recognition module 220 is further adapted to:

[0052] Performing voice preprocessing on the voice activity information of the teacher's recording file and the student's recording file; the voice preprocessing includes frequency division, removal of DC offset and / or pre-emphasis;

[0053] Extract the speech features of each audio frame in the teacher's recording file and the student's recording file after speech preprocessing based on the VAD algorithm; the speech features include short-time energy, zero-crossing rate and / or spectral entropy;

[0054] According to the extracted speech features of each audio frame, it is determined in turn whether each audio frame in the teacher's recording file and the student's recording file contains speech activity information in a loop; if it contains speech activity information, the result corresponding to the audio frame is marked as the first character; if it does not contain speech activity information, the result corresponding to the audio frame is marked as the second character; until the loop completes the determination of the speech features of all audio frames, the first speech string of the teacher's recording file and the second speech string of the student's recording file are obtained.

[0055] Optionally, the speech recognition module 220 is further adapted to:

[0056] It is determined whether the speech feature of each audio frame is greater than a preset speech activity threshold; if so, it is determined that the audio frame contains speech activity information; if not, it is determined that the audio frame does not contain speech activity information.

[0057] Optionally, the speech recognition module 220 is further adapted to:

[0058] The speech features of each audio frame are classified based on a preset classification algorithm to obtain audio frames containing speech activity information and audio frames not containing speech activity information.

[0059] Optionally, the calculation module 230 is further adapted to:

[0060] The first voice string and the second voice string are divided according to a preset length to obtain a plurality of first voice substrings and a plurality of second voice substrings; wherein the number of substrings of the first voice substring and the second voice substring is the same;

[0061] Counting and obtaining first data containing the first character in each first voice substring and second data containing the first character in each second voice substring;

[0062] The ratio of the second data to the sum is calculated based on the sum of the first data and the second data. The average ratio is calculated based on the ratio and the number of substrings as the student opening rate in the online class.

[0063] Optionally, the acquisition module 210 is further adapted to:

[0064] Obtain the recording files of different stages of the online class to calculate the students' speaking rate at each stage.

[0065] The description of each module above refers to the corresponding description in the method embodiment and will not be repeated here.

[0066] An embodiment of the present invention further provides a non-volatile computer storage medium, which stores at least one executable instruction, and the executable instruction can execute operations corresponding to the aperture ratio calculation method based on the VAD algorithm in any of the above method embodiments.

[0067] An embodiment of the present application provides a computer program product, which includes at least one executable instruction or computer program, and the executable instruction or computer program can enable a processor to perform operations corresponding to the aperture ratio calculation method based on the VAD algorithm in any of the above method embodiments.

[0068] Figure 3 A schematic diagram of the structure of a computing device according to an embodiment of the present invention is shown, and the specific implementation of the embodiment of the present invention does not limit the specific implementation of the computing device.

[0069] like Figure 3 As shown, the computing device may include: a processor 302 , a communication interface 304 , a memory 306 , and a communication bus 308 .

[0070] in:

[0071] The processor 302 , the communication interface 304 , and the memory 306 communicate with each other via the communication bus 308 .

[0072] The communication interface 304 is used to communicate with other devices such as clients or other servers.

[0073] The processor 302 is used to execute the program 310, and specifically can execute the relevant steps in the embodiment of the aperture ratio calculation method based on the VAD algorithm.

[0074] Specifically, the program 310 may include program codes, which include computer operation instructions.

[0075] The processor 302 may be a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present invention. The one or more processors included in the computing device may be processors of the same type, such as one or more CPUs; or processors of different types, such as one or more CPUs and one or more ASICs.

[0076] The memory 306 is used to store the program 310. The memory 306 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0077] The program 310 can be specifically used to enable the processor 302 to execute the aperture ratio calculation method based on the VAD algorithm in any of the above-mentioned method embodiments. The specific implementation of each step in the program 310 can refer to the corresponding descriptions in the corresponding steps and units in the above-mentioned aperture ratio calculation embodiment based on the VAD algorithm, which will not be repeated here. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working process of the above-described devices and modules can refer to the corresponding process description in the above-mentioned method embodiment, which will not be repeated here.

[0078] The algorithm or display provided herein is not inherently related to any particular computer, virtual system or other device. Various general purpose systems can also be used together with the teachings based on this. According to the above description, it is obvious that the structure required for constructing such systems. In addition, the embodiment of the present invention is not directed to any specific programming language either. It should be understood that various programming languages ​​can be utilized to realize the content of the embodiment of the present invention described herein, and the description of the above specific language is to disclose the preferred implementation of the embodiment of the present invention.

[0079] In the description provided herein, a large number of specific details are described. However, it is understood that embodiments of the present invention can be practiced without these specific details. In some instances, well-known methods, structures and techniques are not shown in detail so as not to obscure the understanding of this description.

[0080] Similarly, it should be understood that in order to streamline the embodiments of the present invention and to aid in understanding one or more of the various inventive aspects, in the above description of the exemplary embodiments of the present invention, the various features of the embodiments of the present invention are sometimes grouped together into a single embodiment, figure, or description thereof. However, the disclosed method should not be interpreted as reflecting the following intention: that the claimed embodiments of the present invention require more features than the features explicitly recited in each claim. More specifically, as reflected in the claims below, the inventive aspects lie in less than all the features of the single embodiment disclosed above. Therefore, the claims that follow the specific embodiment are hereby expressly incorporated into the specific embodiment, wherein each claim itself serves as a separate embodiment of the present invention.

[0081] Those skilled in the art will appreciate that the modules in the devices in the embodiments may be adaptively changed and arranged in one or more devices different from the embodiments. The modules or units or components in the embodiments may be combined into one module or unit or component, and in addition they may be divided into a plurality of submodules or subunits or subcomponents. Except that at least some of such features and / or processes or units are mutually exclusive, all features disclosed in this specification (including the accompanying claims, abstracts and drawings) and all processes or units of any method or device disclosed in this manner may be combined in any combination. Unless otherwise expressly stated, each feature disclosed in this specification (including the accompanying claims, abstracts and drawings) may be replaced by an alternative feature providing the same, equivalent or similar purpose.

[0082] In addition, those skilled in the art will appreciate that, although some embodiments herein include certain features included in other embodiments but not other features, the combination of features of different embodiments is meant to be within the scope of the present invention and form different embodiments. For example, in the claims below, any one of the claimed embodiments may be used in any combination.

[0083] The various component embodiments of the present invention may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. It should be understood by those skilled in the art that a microprocessor or digital signal processor (DSP) may be used in practice to implement some or all of the functions of some or all of the components according to the embodiments of the present invention. The embodiments of the present invention may also be implemented as a device or apparatus program (e.g., a computer program and a computer program product) for executing part or all of the methods described herein. Such a program implementing an embodiment of the present invention may be stored on a computer-readable medium, or may have the form of one or more signals. Such a signal may be downloaded from an Internet website, or provided on a carrier signal, or provided in any other form.

[0084] It should be noted that the above embodiments illustrate rather than limit the present invention, and that those skilled in the art may devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference symbol between brackets shall not be construed as a limitation on the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "one" or "an" preceding an element does not exclude the presence of a plurality of such elements. The present invention may be implemented by means of hardware comprising a number of different elements and by means of a suitably programmed computer. In a unit claim listing a number of devices, several of these devices may be embodied by the same hardware item. The use of the words first, second, and third, etc. does not indicate any order. These words may be interpreted as names. The steps in the above embodiments, unless otherwise specified, should not be understood as limitations on the order of execution.

Claims

1. A method for calculating aperture ratio based on VAD algorithm, characterized in that: include: Get the recordings of online classes; The recording files include teacher recording files and student recording files; Based on the VAD algorithm, the voice activity information of the teacher's recording file and the student's recording file are detected at preset time intervals, and a first voice string of the teacher's recording file and a second voice string of the student's recording file are obtained; wherein the first voice string and the second voice string are composed of preset characters representing the voice activity information; The voice activity information in the first voice string and the second voice string is counted, and the student opening rate in the online class is calculated based on the statistical results; the student opening rate is determined based on the ratio of the voice activity information in the second voice string to the first voice string.

2. The method according to claim 1, characterized in that The preset characters include a first character representing information containing voice activity and a second character representing information not containing language activity; The method of detecting the voice activity information of the teacher recording file and the student recording file respectively at preset time intervals based on the VAD algorithm to obtain the first voice string of the teacher recording file and the second voice string of the student recording file further includes: Performing voice preprocessing on the voice activity information of the teacher's recording file and the student's recording file; the voice preprocessing includes frequency division, removal of DC offset and / or pre-emphasis; Extracting speech features of each audio frame in the teacher recording file and the student recording file after speech preprocessing based on the VAD algorithm; the speech features include short-time energy, zero-crossing rate and / or spectral entropy; According to the extracted speech features of each audio frame, it is determined in sequence whether each audio frame in the teacher's recording file and the student's recording file contains speech activity information in a loop; if it contains speech activity information, the result corresponding to the audio frame is marked as the first character; if it does not contain speech activity information, the result corresponding to the audio frame is marked as the second character; until the loop completes the determination of the speech features of all audio frames, and obtains the first speech string of the teacher's recording file and the second speech string of the student's recording file.

3. The method according to claim 2, characterized in that The step of sequentially and cyclically determining whether each audio frame in the teacher recording file and the student recording file contains voice activity information based on the extracted voice features of each audio frame further includes: It is determined whether the speech feature of each audio frame is greater than a preset speech activity threshold; if so, it is determined that the audio frame contains speech activity information; if not, it is determined that the audio frame does not contain speech activity information.

4. The method according to claim 2, characterized in that: The step of sequentially and cyclically determining whether each audio frame in the teacher recording file and the student recording file contains voice activity information based on the extracted voice features of each audio frame further includes: The speech features of each audio frame are classified based on a preset classification algorithm to obtain audio frames containing speech activity information and audio frames not containing speech activity information.

5. The method according to claim 2, characterized in that: The counting of the voice activity information in the first voice string and the second voice string, and calculating the student opening rate of the online class according to the statistical results, further comprises: The first voice string and the second voice string are divided according to a preset length to obtain a plurality of first voice substrings and a plurality of second voice substrings; wherein the number of substrings of the first voice substring and the second voice substring is the same; Counting and obtaining first data containing the first character in each first voice substring and second data containing the first character in each second voice substring; The ratio of the second data to the sum is calculated based on the sum of the first data and the second data and the second data. The average ratio is calculated based on the ratio and the number of substrings as the student opening rate in the online class.

6. The method according to any one of claims 1 to 5, characterized in that The obtaining of the recording file of the online class further includes: Obtain the recording files of different stages of the online class to calculate the students' speaking rate at each stage.

7. An aperture ratio calculation device based on VAD algorithm, characterized in that: The device includes: An acquisition module, adapted to acquire recording files of online classes; the recording files include teacher recording files and student recording files; A speech recognition module, adapted to detect the speech activity information of the teacher recording file and the student recording file respectively at preset time intervals based on a VAD algorithm, and obtain a first speech string of the teacher recording file and a second speech string of the student recording file; wherein the first speech string and the second speech string are composed of preset characters representing the speech activity information; The calculation module is suitable for counting the voice activity information in the first voice string and the second voice string, and calculating the student opening rate in the online class according to the statistical results; the student opening rate is determined according to the ratio of the voice activity information in the second voice string to the first voice string.

8. A computing device, characterized in that include: A processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus; The memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform operations corresponding to the aperture ratio calculation method based on the VAD algorithm as described in any one of claims 1 to 6.

9. A computer storage medium, characterized in that The storage medium stores at least one executable instruction, and the executable instruction enables the processor to perform operations corresponding to the aperture ratio calculation method based on the VAD algorithm as described in any one of claims 1 to 6.

10. A computer program product, comprising at least one executable instruction, wherein the executable instruction enables a processor to perform operations corresponding to the method for calculating aperture ratio based on a VAD algorithm according to any one of claims 1 to 6.

Images