Multifunctional network class recorded audio optimizing device

Abstract

The invention relates to a multifunctional network class recorded audio optimization device, and the device comprises an audio signal collection system which is used for collecting audio signals in a teaching environment; the audio preprocessing system is used for converting the analog audio signal into a digital signal and performing gain adjustment; the audio optimization system is used for carrying out noise suppression, echo cancellation and dynamic range control on the audio signal; the voice conversion system is used for converting the optimized audio signals into texts and generating real-time subtitles; the audio data transmission system is used for dynamically selecting a coding and decoding protocol according to a network condition and transmitting audio data; the data secrecy and protection system is used for carrying out encryption and access control on the audio data; and the information display and interaction system is used for providing a visual teaching interface and supporting a teacher-student interaction function. The device can improve audio quality, enhance teaching interactivity and guarantee data safety.

Description

Technical Field

[0001] The present invention relates to the field of online education technology, and in particular to a multifunctional online course recording audio optimization device. Background Art

[0002] With the rapid development of information technology and the widespread adoption of internet infrastructure, online education has become an indispensable component of the modern education system. As a key technology platform supporting online education, online course recording systems integrate multiple technical modules, including audio acquisition, video recording, data processing, and network transmission, enabling the digital recording and dissemination of teaching content through various terminal devices. These systems have been widely adopted in a variety of fields, including higher education institutions, vocational training platforms, corporate in-house training systems, and distance education programs. They have greatly expanded the reach of educational resources, reduced the time and space costs of learning, and provided unprecedented convenience for knowledge dissemination.

[0003] However, in actual application, existing online course recording systems still face many technical bottlenecks and practical challenges in audio processing. The primary problem lies in the audio acquisition and processing links. In typical teaching environments, such as classrooms, conference rooms, or home learning spaces, audio signals are often affected by a variety of interference factors. These interferences include steady-state noise generated by air conditioners and equipment fans, as well as sudden non-steady-state noise such as people walking around, doors and windows opening and closing, and reverberation and echo caused by indoor acoustic reflections. Especially in large classrooms or places with poor acoustic conditions, these interference factors will significantly reduce the signal-to-noise ratio of the voice signal, resulting in a decrease in the clarity of the recorded audio, which seriously affects students' listening experience. Traditional audio processing technologies usually use fixed filtering parameters and noise reduction thresholds, which are difficult to dynamically adjust to different environments and lack the necessary adaptive capabilities and intelligent processing mechanisms.

[0004] Secondly, existing systems have significant shortcomings in interactive functionality. Most traditional online course platforms offer only basic live streaming and recording capabilities, with interaction limited to text chat or simple hand-raising, failing to meet the demands of diverse teaching scenarios. This single interactive model severely restricts teaching effectiveness in teaching sessions requiring real-time Q&A, group discussions, quizzes, or practical demonstrations. Research shows that online courses lacking effective interaction often have student engagement and knowledge retention rates over 30% lower than offline courses. This limitation is particularly pronounced in courses requiring frequent interaction, such as language and arts courses.

[0005] The third key issue concerns data security and privacy protection. The data processed by online course systems contains a large amount of sensitive information, such as the intellectual property of teaching content, student personal information, and online exam data. Existing systems typically rely on basic data encryption transmission and simple permission management mechanisms, which are insufficient to cope with increasingly complex cybersecurity threats. In recent years, cyberattacks targeting educational platforms have become frequent, including incidents such as the theft of course content, intrusions into exam systems, and the leakage of teacher and student information. Especially in cross-border online education scenarios, data protection regulations in different countries and regions have placed more stringent requirements on systems, making traditional data protection solutions difficult to meet actual needs.

[0006] From a technical implementation perspective, existing solutions have three main limitations: in terms of noise processing, most systems use a single spectral subtraction or Wiener filtering algorithm, which is poorly effective in processing non-stationary noise; in terms of echo cancellation, traditional algorithms have poor adaptability to time-varying acoustic environments and are prone to residual echoes or speech distortion; in terms of speech recognition, the recognition accuracy of conventional algorithms often drops significantly in noisy environments, especially in multi-person discussion scenarios, and the effects of speaker separation and speech enhancement need to be improved urgently.

[0007] Furthermore, as online education evolves towards immersive and intelligent learning, users are placing higher demands on systems. The introduction of new technologies such as 4K / 8K ultra-high-definition video, spatial audio, and virtual reality requires audio processing systems to support higher sampling rates and more complex processing algorithms. Furthermore, the addition of innovative features such as real-time multilingual translation, intelligent subtitle generation, and voice sentiment analysis poses new challenges to system computing power and algorithm optimization.

[0008] In summary, developing an online course audio optimization system that can intelligently adapt to complex acoustic environments, provide a rich interactive experience, and have strong security protection capabilities has become a key technical issue that needs to be urgently solved in the field of online education. Summary of the Invention

[0009] The purpose of the present invention is to provide a multifunctional online course recording audio optimization device, which can improve audio quality, enhance teaching interactivity and ensure data security.

[0010] In order to achieve the above-mentioned purpose, the technical solution adopted by the present invention is: a multifunctional online course recording audio optimization device, comprising:

[0011] Audio signal acquisition system, used to collect audio signals in the teaching environment;

[0012] Audio pre-processing system, used to convert analog audio signals into digital signals and perform gain adjustment;

[0013] Audio optimization system for noise suppression, echo cancellation and dynamic range control of audio signals;

[0014] A speech conversion system that converts optimized audio signals into text and generates real-time subtitles;

[0015] Audio data transmission system, used to dynamically select codec protocols and transmit audio data according to network conditions;

[0016] Data confidentiality and protection system for encrypting and controlling access to audio data;

[0017] Information display and interaction system, used to provide a visual teaching interface and support teacher-student interaction functions.

[0018] Furthermore, the audio signal acquisition system includes:

[0019] Microphone array module, used to capture the teacher's voice and ambient noise in all directions;

[0020] A preamplifier module is used to amplify the audio signal to a level suitable for processing;

[0021] Bandpass filter module, used to filter out high-frequency noise and low-frequency interference.

[0022] Furthermore, the audio preprocessing system includes:

[0023] an analog-to-digital converter (ADC) module for converting analog audio signals into digital signals;

[0024] The gain control module is used to automatically adjust the amplitude of the audio signal to avoid volume distortion.

[0025] Furthermore, the audio optimization system includes:

[0026] The speech enhancement module uses hybrid beamforming technology that combines the MVDR (minimum variance distortionless response) algorithm and the GSC (generalized sidelobe canceller) algorithm to suppress steady-state and non-steady-state noise;

[0027] The echo cancellation module uses an adaptive filter to eliminate echoes based on the improved Normalized Least Mean Square Error (NLMS) algorithm;

[0028] The dynamic range control module uses compression expansion technology and automatic gain control algorithm to balance the high and low frequency dynamic range of the audio signal.

[0029] Furthermore, the voice enhancement module further includes:

[0030] A short-time Fourier transform (STFT) unit for converting audio signals from the time domain to the frequency domain;

[0031] Noise covariance matrix calculation unit, used to estimate noise characteristics;

[0032] a steering vector calculation unit, for determining the direction of the target speech signal;

[0033] Adaptive noise cancellation unit is used to further suppress residual noise.

[0034] Furthermore, the voice conversion system includes:

[0035] Speech recognition module, which converts audio signals into text based on deep learning algorithms;

[0036] A real-time subtitle generation module for displaying recognized text as synchronized subtitles;

[0037] Machine translation module for translating subtitles into multiple languages.

[0038] Furthermore, the audio data transmission system supports multiple codec protocols, including AAC, MP3 and Opus, and can dynamically switch codec modes according to network conditions to optimize transmission quality.

[0039] Furthermore, the data confidentiality and protection system includes:

[0040] Data encryption module, used to encrypt audio data stored statically and transmitted dynamically;

[0041] Access control module, used to restrict data access based on user permissions;

[0042] Privacy settings module, used to allow users to customize the scope of data sharing.

[0043] Furthermore, the information display and interaction system includes:

[0044] Visual teaching interface, used to display course content, subtitles and translation results;

[0045] Interactive function module, supporting speech application, voting and online testing;

[0046] The scene mode switching module provides teaching demonstration mode, student interaction mode and online test mode.

[0047] The present invention solves the problems of poor audio processing, lack of interactive functions, and insufficient data security in existing online course audio optimization devices, providing a comprehensive and efficient multifunctional online course recording audio optimization device. Compared with the existing technology, the present invention has the following advantages:

[0048] 1. Wideband noise suppression: Adopting the MVDR+GSC hybrid beamforming algorithm, the main channel achieves optimal frequency domain weighting through the MVDR algorithm, supplemented by the adaptive noise cancellation of the GSC algorithm, to achieve full-band noise suppression within the 300-4000Hz main voice frequency band.

[0049] 2. Intelligent echo cancellation: A deep echo cancellation network is constructed based on the improved NLMS algorithm. The linear echo and nonlinear distortion are processed separately through a double-layer filtering structure, solving the problem of performance degradation of traditional methods in time-varying acoustic environments.

[0050] 3. Optimizing Audio Dynamic Range: The dynamic range control module plays a key role. Its compression and expansion technology and automatic gain control algorithm intelligently adjust the dynamic range of the audio signal. When the audio signal amplitude is too large, it immediately compresses the amplitude to avoid distortion or clipping. When the audio signal amplitude is too small, it quickly expands the amplitude to ensure that even faint sounds are clearly audible.

[0051] 4. Rich interactive functions: Provide real-time translation, voice-to-text and other functions to enhance the interactivity of online courses, improve teaching efficiency and students' learning enthusiasm.

[0052] 5. Comprehensive security and privacy protection: Through functions such as data encryption, we ensure the security of online course data and the privacy of teachers and students, and enhance user trust. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 This is a block diagram of the implementation principle of the multifunctional online course recording audio optimization device provided by an embodiment of the present invention;

[0054] Figure 2 This is a block diagram of the implementation principle of the audio signal acquisition system in an embodiment of the present invention;

[0055] Figure 3 This is a block diagram of the implementation principle of the audio optimization system in an embodiment of the present invention;

[0056] Figure 4 is a workflow diagram of the speech enhancement module in an embodiment of the present invention;

[0057] Figure 5 This is a time domain waveform comparison diagram in an embodiment of the present invention;

[0058] Figure 6 Spectrogram comparison diagram in an embodiment of the present invention;

[0059] Figure 7 This is a block diagram of the implementation principle of the voice conversion system in an embodiment of the present invention. DETAILED DESCRIPTION

[0060] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0061] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present application belongs.

[0062] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or combinations thereof.

[0063] During online course recording, audio signals can be subject to various interferences, impacting students' learning experience. Traditional devices have limited ability to handle these issues. This invention optimizes audio by integrating multiple functional systems that work together.

[0064] like Figure 1 As shown, this embodiment provides a multifunctional online course recording audio optimization device, including: an audio signal acquisition system, an audio preprocessing system, an audio optimization system, a voice conversion system, an audio data transmission system, a data confidentiality and protection system, and an information display and interaction system.

[0065] The audio signal acquisition system is used to collect audio signals in a teaching environment. Figure 2 As shown, the audio signal acquisition system includes a microphone array module, a preamplifier module, and a bandpass filter module. The microphone array module is used to capture the teacher's voice and ambient noise in all directions. The preamplifier module is used to amplify the audio signal to a level suitable for processing. The bandpass filter module is used to filter out high-frequency noise and low-frequency interference. The audio signal acquisition system uses a high-sensitivity array microphone, paired with a preamplifier module and a bandpass filter module, to accurately capture the teacher's voice while effectively suppressing ambient noise and low-frequency interference, ensuring the acquisition of clear original audio signals.

[0066] The audio preprocessing system is used to convert analog audio signals into digital signals and perform gain adjustment. The audio preprocessing system includes an analog-to-digital converter (ADC) module and a gain control module. The analog-to-digital converter (ADC) module is used to convert analog audio signals into digital signals. The gain control module is used to automatically adjust the amplitude of the audio signal to avoid volume distortion. The audio preprocessing system converts analog audio signals into digital signals. The analog-to-digital converter (ADC) module is responsible for signal conversion, and the gain control module performs gain adjustment on the audio signal to prevent the volume from being too high or too low and affecting the recording effect, providing a stable audio foundation for subsequent optimization processing.

[0067] The audio optimization system is used to perform noise suppression, echo cancellation and dynamic range control on the audio signal. Figure 3 As shown, the audio optimization system includes a speech enhancement module, an echo cancellation module and a dynamic range control module. The audio optimization system uses advanced algorithms to deeply optimize the audio after preliminary processing. The speech enhancement module adopts the MVDR+GSC algorithm, combined with microphone array technology, to detect and suppress non-stationary and steady-state noise in real time, highlighting the teacher's voice. The echo cancellation module uses an adaptive filter based on the normalized minimum mean square error (NLMS) algorithm to effectively reduce the echo caused by sound reflection. The dynamic range control module balances the high and low frequency signals of the audio through compression and expansion technology and automatic gain control algorithm, thereby improving the fullness and clarity of the voice.

[0068] The speech enhancement module adopts a hybrid beamforming technology that combines the MVDR (minimum variance distortionless response) algorithm and the GSC (generalized sidelobe canceller) algorithm to suppress steady-state and non-steady-state noise. The speech enhancement module further includes a short-time Fourier transform (STFT) unit, a noise covariance matrix calculation unit, a steering vector calculation unit and an adaptive noise cancellation unit. The short-time Fourier transform (STFT) unit is used to convert the audio signal from the time domain to the frequency domain. The noise covariance matrix calculation unit is used to estimate the noise characteristics. The steering vector calculation unit is used to determine the direction of the target speech signal. The adaptive noise cancellation unit is used to further suppress residual noise. The echo cancellation module eliminates echoes through an adaptive filter based on an improved normalized minimum mean square error (NLMS) algorithm. The dynamic range control module adopts compression and expansion technology and an automatic gain control algorithm to balance the high and low frequency dynamic range of the audio signal.

[0069] The speech conversion system is used to convert the optimized audio signal into text and generate real-time subtitles. Figure 7As shown, the speech conversion system includes a speech recognition module, a real-time subtitle generation module, and a machine translation module. The speech recognition module converts audio signals into text based on a deep learning algorithm. The real-time subtitle generation module is used to display the recognized text as synchronized subtitles. The machine translation module is used to translate the subtitles into multiple languages. The speech conversion system uses deep learning speech recognition methods to quickly and accurately convert recorded speech into text. The real-time subtitle generation and translation module instantly displays the recognized text as subtitles. Through the machine translation (MT) system, multilingual real-time translation is achieved, making it easier for students from different language backgrounds to understand the course content.

[0070] The audio data transmission system is used to dynamically select a codec protocol and transmit audio data based on network conditions. It supports multiple codecs, including AAC, MP3, and Opus, and can dynamically switch codecs based on network conditions to optimize transmission quality. This ensures that processed audio signals are transmitted to students smoothly and with low latency, recreating a realistic classroom interactive experience.

[0071] The data confidentiality and protection system is used to encrypt and control access to audio data. It utilizes multi-layer encryption technology, strict access control, and flexible privacy settings to ensure the security of teaching audio content and user data, prevent data leaks, and protect the privacy of teachers and students. The system includes a data encryption module, an access control module, and a privacy settings module. The data encryption module encrypts audio data for static storage and dynamic transmission. The access control module restricts data access based on user permissions. The privacy settings module allows users to customize the scope of data sharing.

[0072] The information display and interaction system is used to provide a visual teaching interface and support teacher-student interaction functions. The information display and interaction system includes a visual teaching interface, an interactive function module and a scene mode switching module. The visual teaching interface is used to display course content, subtitles and translation results. The interactive function module supports speech applications, voting and online tests. The scene mode switching module provides teaching demonstration mode, student interaction mode and online test mode. The information display and interaction system provides a visual teaching interface to intuitively display classroom content and translation results to students. It also supports interactive functions such as speech applications and voting, and has a scene mode switching function. It can flexibly select teaching demonstration mode, student interaction mode or online test mode according to teaching needs to improve teaching effects and learning experience.

[0073] In online course teaching, the specific implementation method of the multifunctional online course recording audio optimization device provided in this embodiment is as follows.

[0074] Determine device parameters and preparation: Before using the device, clarify the key parameters of each module. Determine the microphone array layout and sampling frequency for the audio signal acquisition system. Select the codec protocol for the audio data transmission system based on the teaching scenario and device performance. Set the language and dictionary type for the speech conversion system based on the course content and student population. Ensure that all hardware devices are properly connected and the software system is running stably. Initialize the data confidentiality and protection system, configuring encryption keys and access permissions.

[0075] Audio signal acquisition stage: The audio signal acquisition system starts working, and the microphone array module captures classroom sounds in real time. Multiple microphones work together to collect the teacher's voice and environmental sounds in all directions, improving signal capture accuracy. The preamplifier module receives the microphone output signal and amplifies it to a level suitable for subsequent circuit processing without affecting the sound quality. The bandpass filter module filters the amplified signal to remove high-frequency noise and low-frequency interference, preliminarily purifying the audio signal and providing a cleaner input for subsequent processing. The audio acquisition system structure diagram is shown below. Figure 2 shown.

[0076] Audio preprocessing: The audio preprocessing system's analog-to-digital converter (ADC) module converts the collected analog audio signals into digital signals for subsequent digital signal processing. The gain control module adjusts the volume of the digital audio signal, automatically boosting low volumes and reducing high volumes based on preset thresholds to ensure the audio signal amplitude remains within a reasonable range, preventing abnormal volume levels from affecting recording and playback.

[0077] Audio optimization stage: The audio optimization system is started. The audio optimization system structure is as follows: Figure 3 shown.

[0078] The speech enhancement module uses the MVDR+GSC algorithm, with the MVDR (Minimum Variance Distortionless Response) algorithm used as the main channel in the GSC (Generalized Sidelobe Canceller) algorithm. First, the audio signal is converted from the time domain to the frequency domain through the Short-Time Fourier Transform (STFT).

[0079] Short-time Fourier transform (STFT):

[0080]

[0081] Among them, X K (m) is the short-time Fourier transform result of the m-th frame and the k-th frequency component, x(n) is the input audio signal, ω(n) is the window function, R is the frame shift length, and N is the number of FFT points.

[0082] For each microphone signal, STFT processing is performed separately to obtain the frequency domain signal matrix. In the frequency domain, the noise covariance matrix is calculated, usually using the first few frames of the signal as the noise reference.

[0083] Noise covariance matrix R nn Calculation:

[0084]

[0085] Where X(f,n) is the frequency domain signal matrix. γ is a small regularization factor used to prevent the matrix from being singular. I is the identity matrix.

[0086] Next, the steering vector is calculated according to the direction of the target signal, and the weight is calculated using the MVDR algorithm.

[0087] Steering vector calculation:

[0088]

[0089] Where f is the frequency, d is the array element spacing, c is the speed of sound, M is the number of microphones, and θtarget is the direction of the target signal.

[0090] MVDR weight calculation:

[0091] For each frequency, calculate the MVDR weight:

[0092]

[0093] Among them, R nn is the noise covariance matrix, and a(f) is the steering vector.

[0094] By applying these weights to the frequency domain signal, an enhanced signal is obtained. Finally, the processed frequency domain signal is converted back to the time domain through the inverse short-time Fourier transform (ISTFT) to obtain the output signal of the main channel. At the same time, a blocking matrix is designed to generate a reference noise signal. The function of the blocking matrix is to eliminate the part of the reference signal associated with the desired signal, thereby extracting a purer noise signal. The reference noise signal is obtained by multiplying the blocking matrix with the original signal and processed in the frequency domain. Subsequently, the reference noise signal is converted back to the time domain through ISTFT. In the time domain, an adaptive noise cancellation algorithm (such as the NLMS algorithm) is used to further suppress the noise. The NLMS algorithm adjusts the weight of the filter so that the output of the filter is as close to the reference noise signal as possible, thereby subtracting the noise component from the main channel signal to obtain an enhanced audio signal. This GSC algorithm, combined with the structure of the MVDR main channel, can effectively suppress noise in a complex noise environment while maintaining the integrity of the desired signal. The implementation process is as follows Figure 4 shown.

[0095] In order to verify the superiority of this algorithm, a simulation experiment was conducted. The classroom was 8 meters long, 6 meters wide, and 3 meters high. The sampling rate was set to 16000kHz. A six-microphone linear array was used, and the spacing between each microphone was 5cm. The sound of the teacher's lecture was first recorded, intercepted for 3 seconds, and low-frequency air-conditioning noise, students whispering (non-steady state), and the impact of moving tables and chairs (transient) were added. Then, reverberation noise was added, and finally the voice signal received by the synthesized microphone was synthesized. After optimization processing by this algorithm, the STOI of the voice signal increased from 0.372 to 0.540, and the SNR increased from -26.00dB to -1.56dB. The time domain waveform comparison is shown in the figure below. Figure 5 As shown in the spectrogram comparison chart, Figure 6 shown.

[0096] The echo cancellation module uses an adaptive filter based on the Normalized Least Mean Square (NLMS) algorithm, which iteratively updates the filter coefficients to minimize the error between the echo signal and the desired signal. The specific calculation formula is as follows:

[0097] The filter output is:

[0098] y(n)=w T (n)x(n)

[0099] Where y(n) is the output of the filter, w T (n) is the filter coefficient vector, and x(n) is the input signal vector.

[0100] Error signal:

[0101] e(n)=d(n)-y(n)

[0102] Where e(n) is the error signal and d(n) is the desired signal.

[0103] The coefficient update formula is:

[0104]

[0105] where μ is the step size factor that controls the update rate; δ is a small regularization constant used to avoid division by zero errors, and ||x(n)|| 2 is the energy of the input signal, usually x(n) T x(n).

[0106] The dynamic range control module uses compression and expansion technology and automatic gain control algorithm to monitor the audio signal amplitude in real time, balance high and low frequency signals, and improve voice fullness and clarity.

[0107] Speech conversion stage: The speech recognition module of the speech conversion system uses deep learning methods to convert the optimized audio signal into an acoustic feature vector. Combined with the probability distribution of the vocabulary sequence provided by the language model module, the search decoding module uses the Viterbi algorithm to convert the audio to text. The real-time subtitle generation and translation module displays the recognized text as subtitles in real time and translates the subtitles into the specified language through the machine translation system, making it easier for students from different language backgrounds to understand the course content. The structure of the speech conversion system is as follows: Figure 7 shown.

[0108] During audio data transmission, the system selects the appropriate audio codec standard and protocol based on network conditions and device compatibility. For example, AAC is preferred for high quality when the network is stable, while Opus is switched to enhance packet loss mitigation during network fluctuations. During transmission, the system continuously monitors network conditions and dynamically adjusts transmission parameters to ensure smooth, low-latency audio signal transmission to students.

[0109] Data confidentiality and protection: The data confidentiality and protection system operates throughout the entire audio data generation, storage, and transmission process. The data encryption module encrypts audio data at rest and in transit, protecting it from theft and tampering. The access control module strictly limits data access levels for different users based on pre-set permissions. The privacy settings function allows users to independently select privacy policies for their personal information and learning data, ensuring user privacy and security.

[0110] Information Display and Interaction Stage: The information display and interaction system visually presents classroom content, translated subtitles, and other content to students. In teaching demonstration mode, teachers present teaching materials, while in interactive mode, students can speak, participate in discussions, and vote. The system supports scene mode switching. In online testing mode, teachers post test questions, students submit their responses, and teachers view and provide feedback in real time, enhancing teaching interactivity and effectiveness.

[0111] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any other manner. Any person skilled in the art may utilize the above-disclosed technical content to modify or modify the present invention into equivalent embodiments. However, any simple modifications, equivalent variations, and modifications to the above embodiments that do not depart from the technical content of the present invention and are based on the technical essence of the present invention remain within the scope of protection of the present invention.

Claims

1. A multifunctional online course recording audio optimization device, characterized in that: include: Audio signal acquisition system, used to collect audio signals in the teaching environment; Audio pre-processing system, used to convert analog audio signals into digital signals and perform gain adjustment; Audio optimization system for noise suppression, echo cancellation and dynamic range control of audio signals; A speech conversion system that converts optimized audio signals into text and generates real-time subtitles; Audio data transmission system, used to dynamically select codec protocols and transmit audio data according to network conditions; Data confidentiality and protection system for encrypting and controlling access to audio data; Information display and interaction system, used to provide a visual teaching interface and support teacher-student interaction functions.

2. The multifunctional online course recording audio optimization device according to claim 1, characterized in that: The audio signal acquisition system comprises: Microphone array module, used to capture the teacher's voice and ambient noise in all directions; A preamplifier module is used to amplify the audio signal to a level suitable for processing; Bandpass filter module, used to filter out high-frequency noise and low-frequency interference.

3. The multifunctional online course recording audio optimization device according to claim 1, characterized in that: The audio preprocessing system comprises: an analog-to-digital converter (ADC) module for converting analog audio signals into digital signals; The gain control module is used to automatically adjust the amplitude of the audio signal to avoid volume distortion.

4. The multifunctional online course recording audio optimization device according to claim 1, characterized in that: The audio optimization system comprises: The speech enhancement module uses hybrid beamforming technology that combines the MVDR (minimum variance distortionless response) algorithm and the GSC (generalized sidelobe canceller) algorithm to suppress steady-state and non-steady-state noise; The echo cancellation module uses an adaptive filter to eliminate echoes based on the improved Normalized Least Mean Square Error (NLMS) algorithm; The dynamic range control module uses compression expansion technology and automatic gain control algorithm to balance the high and low frequency dynamic range of the audio signal.

5. The multifunctional online course recording audio optimization device according to claim 4, characterized in that: The speech enhancement module further comprises: A short-time Fourier transform (STFT) unit for converting audio signals from the time domain to the frequency domain; Noise covariance matrix calculation unit, used to estimate noise characteristics; a steering vector calculation unit, for determining the direction of the target speech signal; Adaptive noise cancellation unit is used to further suppress residual noise.

6. The multifunctional online course recording audio optimization device according to claim 1, characterized in that: The speech conversion system comprises: Speech recognition module, which converts audio signals into text based on deep learning algorithms; A real-time subtitle generation module for displaying recognized text as synchronized subtitles; Machine translation module for translating subtitles into multiple languages.

7. The multifunctional online course recording audio optimization device according to claim 1, characterized in that: The audio data transmission system supports multiple codec protocols, including AAC, MP3 and Opus, and can dynamically switch codec modes according to network conditions to optimize transmission quality.

8. The multifunctional online course recording audio optimization device according to claim 1, characterized in that: The data confidentiality and protection system includes: Data encryption module, used to encrypt audio data stored statically and transmitted dynamically; Access control module, used to restrict data access based on user permissions; Privacy settings module, used to allow users to customize the scope of data sharing.

9. The multifunctional online course recording audio optimization device according to claim 1, characterized in that: The information display and interaction system includes: Visual teaching interface, used to display course content, subtitles and translation results; Interactive function module, supporting speech application, voting and online testing; The scene mode switching module provides teaching demonstration mode, student interaction mode and online test mode.

10. The multifunctional online course recording audio optimization device according to any one of claims 1 to 9, characterized in that: The device is suitable for online education platforms and can improve audio quality, enhance teaching interactivity and ensure data security.

Images