Sound channel correction method and device, equipment and storage medium

Abstract

The invention relates to the technical field of audio equipment correction, and discloses a sound channel correction method and device, equipment and a storage medium, which are used for improving the accuracy of sound channel correction of audio equipment. The sound channel correction method comprises the following steps: acquiring an audio signal of audio equipment and user historical data; identifying a current use scene based on the audio signal; performing sound channel characteristic analysis on the left and right sound channels based on the audio signal; performing user habit analysis and scene adaptability analysis according to the user historical data and the current use scene; and correcting the parameters of the left and right sound channels of the audio equipment according to the sound channel characteristic analysis result, the user habit analysis result and the scene adaptability analysis result.

Description

Technical Field

[0001] The present invention relates to the technical field of audio equipment correction, and in particular to a sound channel correction method, device, equipment and storage medium. Background Art

[0002] The existing technology usually directly relies on the channel characteristics of the audio signal when calibrating the left and right channels of audio equipment, ignoring the user's personalized auditory preferences and habits, as well as the impact of different usage scenarios on the audio output effect, resulting in inaccurate and lack of specificity in the correction results, making it difficult to accurately match the user's actual auditory needs, which in turn causes a poor user experience. Summary of the invention

[0003] The present invention provides a channel correction method, device, equipment and storage medium to solve the problem in the prior art that the channel correction result is not accurate enough due to a single correction.

[0004] A first aspect of the present invention provides a channel correction method, comprising: obtaining an audio signal and user history data of an audio device, the audio signal including information of left and right channels; identifying a current usage scenario based on the audio signal; performing channel characteristic analysis of left and right channels based on the audio signal; performing user habit analysis and scene adaptability analysis based on the user history data and the current usage scenario; and correcting parameters of left and right channels of the audio device based on the channel characteristic analysis results, the user habit analysis results, and the scene adaptability analysis results.

[0005] In a feasible implementation, identifying the current usage scenario based on the audio signal includes: analyzing the ambient noise in the audio signal to obtain a first analysis result; analyzing the voice information in the audio signal to obtain a second analysis result; analyzing the music type in the audio signal to obtain a third analysis result; and performing a weight analysis based on the first analysis result, the second analysis result, and the third analysis result to obtain the current usage scenario.

[0006] In a feasible implementation, the analyzing the ambient noise in the audio signal to obtain a first analysis result includes: extracting noise characteristics of the ambient noise in the audio signal; matching the noise characteristics with a preset noise database; and obtaining a first analysis result based on the matching result, wherein the first analysis result includes at least one usage scenario.

[0007] In a feasible implementation, the analyzing the voice information in the audio signal to obtain a second analysis result includes: performing voice recognition on the audio signal and converting the voice into text information; analyzing the semantic content of the text information and determining the category or topic to which the voice belongs based on the semantic content; determining the second analysis result based on the category or topic, the second analysis result including at least one usage scenario.

[0008] In a feasible implementation, the analyzing the music type in the audio signal to obtain a third analysis result includes: extracting music features in the audio signal; inputting the music features into a preset music classification model to obtain the audio type; determining the music style to which the music type belongs; and determining the third analysis result based on the music style, wherein the third analysis result includes at least one usage scenario.

[0009] In a feasible implementation, the weight analysis is performed based on the first analysis result, the second analysis result, and the third analysis result to obtain the current usage scenario, including: calculating the total weight of each usage scenario based on the first analysis result, the second analysis result, and the third analysis result; and selecting the usage scenario with the highest total weight as the current usage scenario.

[0010] In a feasible implementation, the channel characteristic analysis of the left and right channels based on the audio signal includes: performing frequency response analysis of the left and right channels based on the audio signal; performing phase analysis of the left and right channels based on the audio signal; and performing time delay analysis of the left and right channels based on the audio signal.

[0011] In a feasible implementation, the frequency response analysis of the left and right channels based on the audio signal includes: obtaining frequency response curves of the left and right channels based on the audio signal; determining the gain imbalance of the left and right channels within a specific frequency range using the difference in the frequency response curves; and analyzing whether the frequency response of the left and right channels meets a preset consistency requirement based on the gain imbalance.

[0012] In a feasible implementation, performing phase analysis of the left and right channels based on the audio signal includes: acquiring phase spectra of the left and right channels based on the audio signal; comparing the phase spectra of the left and right channels to identify spectral regions of phase deviation or phase distortion; and analyzing whether the phase response of the left and right channels meets a preset phase requirement based on the degree of the phase deviation or phase distortion.

[0013] In a feasible implementation, the time delay analysis of the left and right channels based on the audio signal includes: measuring the time delay between the left and right channels based on the audio signal; calculating a synchronization index between the left and right channels based on the time delay; and analyzing whether the synchronization of the left and right channels meets a preset delay requirement based on the synchronization index.

[0014] In a feasible implementation, the user habit analysis and scenario adaptability analysis based on the user historical data and the current usage scenario include: extracting user behavior information from the user historical data, the behavior information including audio playback records, device settings and user feedback, and analyzing the user's usage preferences based on the behavior information; determining the scenario preference based on the current usage scenario.

[0015] In a feasible implementation, the parameters of the left and right channels of the audio device are corrected according to the channel characteristics analysis results, the user habits analysis results and the scene adaptability analysis results, including: prioritizing and weighting all parameters according to the degree of influence of preset audio parameters on audio playback; constructing a specific correction strategy for each parameter based on the channel characteristics analysis results, the user habits analysis results and the scene adaptability analysis results; and correcting the parameters of the left and right channels of the audio device according to the correction strategy.

[0016] The second aspect of the present invention provides a channel correction device, including: an acquisition module, used to acquire an audio signal and user history data output by an audio device, the audio signal including information of left and right channels; an identification module, used to identify a current usage scenario based on the audio signal; a first analysis module, used to perform channel characteristic analysis of left and right channels based on the audio signal; a second analysis module, used to perform user habit analysis and scene adaptability analysis based on the user history data and the current usage scenario; and a correction module, used to correct the parameters of the left and right channels of the audio device based on the channel characteristic analysis results, the user habit analysis results and the scene adaptability analysis results.

[0017] In a feasible implementation, the recognition module includes: a first analysis unit, used to analyze the ambient noise in the audio signal to obtain a first analysis result; a second analysis unit, used to analyze the voice information in the audio signal to obtain a second analysis result; a third analysis unit, used to analyze the music type in the audio signal to obtain a third analysis result; and a fourth analysis unit, used to perform a weight analysis based on the first analysis result, the second analysis result and the third analysis result to obtain the current usage scenario.

[0018] In a feasible implementation, the first analysis unit is specifically used to: extract noise characteristics of ambient noise in the audio signal; match the noise characteristics with a preset noise database; and obtain a first analysis result based on the matching result, wherein the first analysis result includes at least one usage scenario.

[0019] In a feasible implementation, the second analysis unit is specifically used to: perform speech recognition on the audio signal and convert the speech into text information; analyze the semantic content of the text information and determine the category or topic to which the speech belongs based on the semantic content; determine a second analysis result based on the category or topic, and the second analysis result includes at least one usage scenario.

[0020] In a feasible implementation, the third analysis unit is specifically used to: extract music features from the audio signal; input the music features into a preset music classification model to obtain an audio type; determine the music style to which the music type belongs; and determine a third analysis result based on the music style, wherein the third analysis result includes at least one usage scenario.

[0021] In a feasible implementation, the fourth analysis unit is specifically used to: calculate the total weight of each usage scenario based on the first analysis result, the second analysis result and the third analysis result; and select the usage scenario with the highest total weight as the current usage scenario.

[0022] In a feasible implementation, the first analysis module includes: a frequency analysis unit, used to perform frequency response analysis of left and right channels based on the audio signal; a phase analysis unit, used to perform phase analysis of left and right channels based on the audio signal; and a time delay analysis unit, used to perform time delay analysis of left and right channels based on the audio signal.

[0023] In a feasible implementation, the frequency analysis unit is specifically used to: obtain frequency response curves of left and right channels based on the audio signal; determine the gain imbalance of left and right channels within a specific frequency range using the difference in the frequency response curves; and analyze whether the frequency response of the left and right channels meets a preset consistency requirement based on the gain imbalance.

[0024] In a feasible implementation, the phase analysis unit is specifically used to: obtain the phase spectrum of the left and right channels based on the audio signal; compare the phase spectra of the left and right channels to identify the spectrum area of ​​the phase deviation or phase distortion; and analyze whether the phase response of the left and right channels meets the preset phase requirement according to the degree of the phase deviation or phase distortion.

[0025] In a feasible implementation, the time delay analysis unit is specifically used to: measure the time delay between the left and right channels based on the audio signal; calculate the synchronization index between the left and right channels according to the time delay; and analyze whether the synchronization of the left and right channels meets the preset delay requirement according to the synchronization index.

[0026] In a feasible implementation, the second analysis module is specifically used to: extract user behavior information from the user historical data, the behavior information including audio playback records, device settings and user feedback, and analyze the user's usage preferences based on the behavior information; determine the scenario preference based on the current usage scenario.

[0027] In a feasible implementation, the correction module is specifically used to: prioritize and weight all parameters according to the degree of influence of preset audio parameters on audio playback; construct a specific correction strategy for each parameter based on the channel characteristic analysis results, the user habit analysis results and the scene adaptability analysis results; and correct the parameters of the left and right channels of the audio device according to the correction strategy.

[0028] The third aspect of the present invention provides a channel correction device, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor calls the instructions in the memory so that the channel correction device executes the above-mentioned channel correction method.

[0029] A fourth aspect of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, which, when executed on a computer, enable the computer to execute the above-mentioned vocal tract correction method.

[0030] In the technical solution provided by the present invention, an audio signal and user history data of an audio device are obtained, the audio signal including information of two left and right channels; the current usage scenario is identified based on the audio signal; channel characteristics analysis of the left and right channels is performed based on the audio signal; user habit analysis and scene adaptability analysis are performed according to the user history data and the current usage scenario; and parameters of the left and right channels of the audio device are corrected according to the channel characteristics analysis results, the user habit analysis results, and the scene adaptability analysis results. In an embodiment of the present invention, the left and right channel parameters of the audio device are accurately corrected by comprehensively considering the channel characteristics of the audio signal, the personalized hearing habits of the user, and the specific usage scenarios, which significantly improves the pertinence and accuracy of the channel correction of the audio device, thereby optimizing the user's auditory experience and ensuring that the audio content meets the actual needs of the user in different scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1A schematic diagram of an embodiment of a vocal tract correction method in an embodiment of the present invention;

[0032] Figure 2 A schematic diagram of another embodiment of the vocal tract correction method according to an embodiment of the present invention;

[0033] Figure 3 A schematic diagram of an embodiment of a vocal tract correction device according to an embodiment of the present invention;

[0034] Figure 4 A schematic diagram of another embodiment of the vocal tract correction device in an embodiment of the present invention;

[0035] Figure 5 FIG. 4 is a schematic diagram of an embodiment of a vocal tract correction device in an embodiment of the present invention. DETAILED DESCRIPTION

[0036] The embodiments of the present invention provide a channel correction method, apparatus, device and storage medium, which comprehensively analyze audio signal characteristics, user habits and current usage scenarios, intelligently correct left and right channel parameters of audio equipment, and improve the accuracy of channel correction of audio equipment.

[0037] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0038] It is understandable that the execution subject of the present invention may be a channel correction device, or a terminal or a server, which is not limited here. The embodiment of the present invention is described by taking a server as the execution subject as an example.

[0039] For ease of understanding, the specific process of the embodiment of the present invention is described below. Figure 1 , an embodiment of the vocal tract correction method in the embodiment of the present invention includes:

[0040] 101. Obtain an audio signal and user history data of an audio device, where the audio signal includes information of left and right channels;

[0041] High-precision microphones are used to collect audio signals from audio equipment. High-precision measurement microphones have extremely high sensitivity and accuracy and can capture subtle changes in audio signals, including frequency response, phase response, and sound pressure level.

[0042] The microphone will collect the response data of the left and right channels at the same time. To ensure the synchronization of the data, professional audio acquisition software or hardware systems are usually used to record these signals. In order to improve the accuracy of the measurement, multiple measurements are usually taken and the average is taken, which helps to reduce random errors and uncertainties and improve the reliability of the final results.

[0043] 102. Identify the current usage scenario based on the audio signal;

[0044] Voice activity detection technology (VAD) is used to separate the speech part from the non-speech part of the audio signal, where the non-speech part is the background noise signal. The separated background noise signal is subjected to power spectral density calculation or energy integration to quantify its intensity or energy level. Then, based on the calculated power or energy value and referring to the preset noise level threshold or classification standard, it is evaluated whether the noise level of the current environment is high, medium or low, or a specific decibel (dB) value is given to indicate the noise intensity. A low noise level usually means an indoor or quiet environment, while a high noise level may indicate an outdoor or noisy public transportation environment.

[0045] Preprocess the audio signal to improve signal quality and reduce noise interference; apply spectrum analysis techniques, such as short-time Fourier transform (STFT) or Mel-frequency cepstral coefficients (MFCC), to extract the spectral characteristics of the audio signal, which can reflect the frequency distribution of sound in different scenarios; calculate the rhythm characteristics of the audio signal, including but not limited to zero crossing rate, energy change rate, beat detection, etc., to capture the rhythm of the sound and the dynamic changes of volume;

[0046] In addition, other relevant features can be extracted, such as the duration, peak frequency, and spectrum flatness of the audio signal, to further enrich the feature set and improve the accuracy of scene recognition. Machine learning techniques, such as feature selection algorithms, are used to evaluate and screen the extracted features to retain the most valuable feature data for scene recognition and remove redundant or irrelevant features, thereby optimizing the feature set.

[0047] Based on the above noise level and feature set, machine learning or deep learning algorithms are used to identify the current usage scenario. These algorithms may include classifiers, regressors or clusterers, etc., which can map audio signals to predefined scene categories based on feature data. Scene categories include but are not limited to indoor, outdoor, conference room, concert, etc.

[0048] 103. Analyze the channel characteristics of the left and right channels based on the audio signal;

[0049] In the audio signal analysis, digital filters are used to remove environmental noise and background interference to ensure data accuracy. The time domain signal is converted into a frequency domain signal by performing a fast Fourier transform (FFT) on the signals corresponding to the left and right channels, and a spectrum diagram is plotted to observe the consistency of the frequency response. Phase information is extracted from the FFT results, the phase difference between the left and right channels at each frequency is calculated, and a phase difference-frequency curve is plotted to analyze the phase consistency. In addition, the root mean square (RMS) value of the left and right channel signals is calculated to measure the volume, and the RMS values ​​are compared to calculate the volume balance difference percentage. All these analysis results are summarized into a report, including key indicators such as frequency response curve, phase difference-frequency curve and volume balance difference, to comprehensively evaluate the performance of the audio equipment.

[0050] 104. Conduct user habit analysis and scenario adaptability analysis based on user historical data and current usage scenarios;

[0051] Cluster analysis is used to divide user historical data into different categories of user behavior data. Cluster analysis techniques include at least one of the K-means algorithm and the DBSCAN algorithm. Time series analysis techniques are used to analyze user behavior characteristics in different time periods based on different categories of user behavior data, so as to identify user habits and behavior patterns. Time series analysis techniques include at least one of the ARIMA model and the LSTM model. Behavioral characteristics include identifying user activity and preference changes in different time periods. Association rule mining and collaborative filtering algorithms are used to determine user preferences based on user habits and behavior patterns. Association rule mining algorithms include at least one of the Apriori algorithm and the FP-Growth algorithm. Collaborative filtering algorithms include at least one of user-based collaborative filtering and project-based collaborative filtering.

[0052] Filter out corresponding scene preferences from the scene-audio analysis library according to the current usage scenario.

[0053] 105. Correct the parameters of the left and right channels of the audio device according to the results of the channel characteristic analysis, the user habit analysis, and the scene adaptability analysis.

[0054] Identify the difference between the left and right channels based on the frequency response curve, and calculate the gain or attenuation adjustment value corresponding to the frequency segment with significant difference; determine the frequency area with significant phase difference between the left and right channels based on the phase difference-frequency curve, and calculate the corresponding phase compensation parameter; calculate the volume balance adjustment ratio by measuring the volume difference between the left and right channels; calculate the parameters that need to be adjusted for the left and right channels and the corresponding adjustment values ​​based on user preferences; calculate the parameters that need to be adjusted for the left and right channels and the corresponding adjustment values ​​based on scene preferences.

[0055] The final adjustment values ​​of the parameters corresponding to the left and right channels are calculated based on the adjustment values ​​of each parameter in the results of comprehensive channel characteristic analysis, user habit analysis and scene adaptability analysis. The parameters of the left and right channels of the audio device are corrected based on the adjustment values ​​to ensure that the adjustment results can meet user preferences, adapt to the current scene, and maintain the balance of channel characteristics.

[0056] In an embodiment of the present invention, an audio signal and user history data of an audio device are obtained, the audio signal including information of two left and right channels; the current usage scenario is identified based on the audio signal; channel characteristics analysis of the left and right channels is performed based on the audio signal; user habit analysis and scene adaptability analysis are performed based on the user history data and the current usage scenario; and parameters of the left and right channels of the audio device are corrected based on the channel characteristics analysis results, the user habit analysis results, and the scene adaptability analysis results. This solution achieves refined correction of the left and right channel parameters of the audio device by comprehensively considering the channel characteristics of the audio signal, the user's personalized hearing habits, and the specific usage scenarios, thereby improving the accuracy and pertinence of the audio device channel correction, thereby optimizing the user's auditory experience, and ensuring that the audio content meets the actual needs of the user in different scenarios.

[0057] See also Figure 2 Another embodiment of the vocal tract correction method in the embodiment of the present invention includes:

[0058] 201. Obtain an audio signal and user history data of an audio device, where the audio signal includes information of left and right channels;

[0059] 202. Analyze the environmental noise in the audio signal to obtain a first analysis result;

[0060] Extracting noise characteristics of ambient noise in an audio signal; matching the noise characteristics with a preset noise database; and obtaining a first analysis result based on the matching result, the first analysis result including at least one usage scenario.

[0061] The audio data is preprocessed, including removal of DC components, filtering (such as high-pass filtering to remove low-frequency noise, low-pass filtering to remove high-frequency noise), normalization, etc., to improve the accuracy of subsequent processing. The noise characteristics of the ambient noise are extracted from the preprocessed audio signal using an audio feature extraction method. The noise characteristics include statistical characteristics such as mean, variance, autocorrelation function, power spectrum density, and frequency domain characteristics such as spectrum graph, frequency characteristics, and spectrum envelope. It can also include time domain characteristics, such as time domain waveform graph and energy graph, and time-frequency domain characteristics, such as time-frequency graph obtained by short-time Fourier transform (STFT) or continuous wavelet transform (CWT).

[0062] The noise characteristics are compared and analyzed with the preset noise database, which pre-stores the characteristic data of various environmental noises, including but not limited to urban traffic noise, indoor life noise, natural sounds (such as wind and rain), industrial machinery noise, etc. Each noise type is accompanied by a detailed characteristic description and usage scenario label.

[0063] Using a similarity or distance measurement algorithm, calculate the target similarity or target distance between the noise characteristics and the characteristic parameters of each noise type in a preset noise database; compare the target similarity or target distance with the corresponding similarity threshold or distance threshold, and determine at least one usage scenario based on the comparison result to obtain a first analysis result, which represents one or more specific usage scenarios that the ambient noise in the audio signal most likely corresponds to.

[0064] According to the actual application requirements and the distribution of noise characteristics, a reasonable similarity threshold or distance threshold is set. If the target similarity exceeds the set similarity threshold, or the target distance is less than the set distance threshold, it is considered that the extracted noise characteristics match the corresponding noise type successfully, and the corresponding usage scenario label is determined based on the successfully matched noise type to obtain at least one usage scenario. Furthermore, a threshold for the number of selected scenarios can also be set. For example, if the threshold for the number of selected scenarios is 2, and there are 4 usage scenario labels corresponding to the noise type, then 2 usage scenarios with higher priority are selected according to the priority of the similarity corresponding to the noise type.

[0065] 203. Analyze the voice information in the audio signal to obtain a second analysis result;

[0066] Perform speech recognition on the audio signal and convert the speech into text information; analyze the semantic content of the text information and determine the category or topic to which the speech belongs based on the semantic content; determine a second analysis result based on the category or topic, and the second analysis result includes at least one usage scenario.

[0067] Extract speech features from audio signals, such as Mel-frequency cepstral coefficients (MFCC), spectrum features, fundamental frequency features, etc., use pre-trained speech recognition models (such as deep learning models, hidden Markov models, etc.) to decode the extracted speech features, and convert the audio signals into text information.

[0068] The selection and training of speech recognition models should be customized and optimized according to the actual application scenarios and language characteristics. The text information obtained by speech recognition is processed by natural language, including word segmentation, part-of-speech tagging, syntactic analysis and other steps to extract key information and semantic structure in the text.

[0069] Semantic understanding technology (such as semantic role labeling, entity recognition, sentiment analysis, etc.) is used to analyze key information and semantic results to obtain semantic content, and the category or topic to which the text information belongs is determined based on the semantic content. For example, classification algorithms (such as support vector machines, decision trees, neural networks, etc.) or topic models (such as LDA, PLSA, etc.) are used to achieve category or topic judgment, and the text information is mapped to the preset usage scenario based on the obtained category or topic.

[0070] 204. Analyze the music type in the audio signal to obtain a third analysis result;

[0071] Extract music features from an audio signal; input the music features into a preset music classification model to obtain an audio type; determine the music style to which the music type belongs; and determine a third analysis result based on the music style, wherein the third analysis result includes at least one usage scenario.

[0072] Music features are extracted from audio signals, and the music features generally include but are not limited to rhythm features, melody features, harmony features, texture features, etc. The feature extraction method can be based on signal processing technology or machine learning methods.

[0073] Music classification models include but are not limited to support vector machines, decision trees, random forests, neural networks, etc. The extracted music features are input into the music classification model to obtain the classification result of the audio signal, that is, the audio type. According to the audio type, it is mapped to the corresponding music style, which can be achieved through a predefined style-type mapping table or a rule-based reasoning system. If the audio signal contains multiple music types or style elements, it is necessary to combine multiple classification results and feature information to judge its overall style, and obtain the third analysis result according to the music style matching corresponding to at least one usage scenario.

[0074] 205. Perform weight analysis based on the first analysis result, the second analysis result, and the third analysis result to obtain a current usage scenario;

[0075] The total weight of each usage scenario is calculated based on the first analysis result, the second analysis result, and the third analysis result; and the usage scenario with the highest total weight is selected as the current usage scenario.

[0076] By matching the ambient noise with the preset noise database, the first analysis result gives two possible usage scenarios: "family leisure" and "children's entertainment", and the corresponding weights are 0.4, 0.3 and 0.2 respectively; because the voice information is captured: "Baby, let's go listen to a new story." Through voice recognition technology, the second analysis result gives two possible usage scenarios: "parent-child reading" and "story time", and the corresponding weights are 0.4 and 0.3; by analyzing the type of music in the audio signal, the third analysis result gives two possible usage scenarios: "children's entertainment" and "parent-child reading", and the corresponding weights are 0.4 and 0.2. The above same scenarios are merged, and the merged results are: the total weight of "family leisure" = 0.4; the total weight of "children's entertainment" = 0.3+0.4=0.7; the total weight of "parent-child reading" = 0.4+0.2=0.6; the total weight of "story time" = 0.3. From the above analysis results, it can be seen that the total weight of "children's entertainment" is the largest, so "children's entertainment" is used as the current usage scenario.

[0077] 206. Perform frequency response analysis of left and right channels based on audio signals;

[0078] Obtain frequency response curves of the left and right channels based on the audio signal; determine the gain imbalance of the left and right channels within a specific frequency range using the difference in the frequency response curves; and analyze whether the frequency response of the left and right channels meets the preset consistency requirements based on the gain imbalance.

[0079] 207. Perform phase analysis of left and right channels based on audio signals;

[0080] The phase spectra of the left and right channels are obtained based on the audio signal; the phase spectra of the left and right channels are compared to identify the spectrum area of ​​phase deviation or phase distortion; and the phase response of the left and right channels is analyzed to see whether it meets the preset phase requirement according to the degree of phase deviation or phase distortion.

[0081] 208. Perform time delay analysis of left and right channels based on the audio signal;

[0082] The time delay between the left and right channels is measured based on the audio signal; the synchronization index between the left and right channels is calculated based on the time delay; and whether the synchronization of the left and right channels meets the preset delay requirement is analyzed based on the synchronization index.

[0083] 209. Conduct user habit analysis and scenario adaptability analysis based on user historical data and current usage scenarios;

[0084] Extract user behavior information from user historical data, including audio playback records, device settings, and user feedback, and analyze user usage preferences based on the behavior information; determine scenario preferences based on the current usage scenario.

[0085] Analyzing the user's usage preferences includes: analyzing whether the user tends to adjust the balance of the left and right channels, and the direction of the adjustment (such as the left channel is heavier or the right channel is heavier), identifying the user's left and right channel balance preferences in different usage scenarios (such as listening to music, watching movies, playing games), analyzing the user's adjustments in different frequency bands, identifying the user's specific needs for sound quality, identifying the user's equalizer setting preferences under different audio types (such as pop music, classical music, jazz), analyzing the frequency and scenarios of the user's selection of specific sound effect modes, understanding the user's preferences for sound effects, identifying the user's sound effect mode selection preferences under different usage needs (such as relaxation, concentration, and exercise), analyzing the user's preferences for parameters such as volume, playback speed, and sound clarity, and identifying changes in the user's audio parameter settings in different time periods or usage scenarios.

[0086] The scene preference is determined based on the current usage scene, and the left and right channel parameter preference settings corresponding to the scene are retrieved from a preset scene preference database to obtain the scene preference.

[0087] 210. Correct parameters of left and right channels of the audio device according to the channel characteristic analysis results, the user habit analysis results, and the scene adaptability analysis results.

[0088] Prioritize and weight all parameters according to their impact on audio playback; build a specific correction strategy for each parameter based on the channel characteristics analysis results, user habit analysis results, and scene adaptability analysis results; correct the parameters of the left and right channels of the audio device according to the correction strategy.

[0089] Identify and analyze key parameters in audio playback, including but not limited to channel balance, frequency response, phase response, volume parameters, and equalizer parameters; evaluate the specific impact of each parameter on audio playback quality, taking into account factors including but not limited to sound clarity, stereo effect, timbre accuracy, and user-perceived comfort; assign a priority to each parameter based on the evaluation results, and sort the priority according to the importance of the parameter's impact on audio playback quality; assign a weight value to each parameter, which reflects the relative importance of the parameter in the overall audio parameter correction process; record and output the results of priority and weight assignment for use as a reference in subsequent audio parameter correction processes. For example: set the channel balance parameter that directly affects the sound positioning and stereo effect to the highest priority; set the frequency response and phase response parameters that affect sound clarity and timbre accuracy to a medium priority; set the volume parameter and equalizer parameter that affect the user's volume preference and sound quality perception to a lower priority, but in specific user scenarios and needs, the priority of these parameters can be adjusted in a timely manner.

[0090] The first audio parameter that needs to be adjusted is calculated based on the frequency response analysis, phase analysis and time delay analysis of the left and right channels, and the first audio parameter includes frequency, phase and time; the second audio parameter that needs to be adjusted is calculated based on the user preference, and the second audio parameter includes but is not limited to volume, sound effects and equalizer, etc.; the third audio parameter that needs to be adjusted is calculated based on the scene preference, and the third audio parameter includes but is not limited to frequency, phase and channel balance, etc.; the adjustment value required for each parameter is calculated based on the first audio parameter, the second audio parameter and the third audio parameter and the corresponding weight of each parameter, and each parameter is adjusted accordingly according to the priority.

[0091] In an embodiment of the present invention, an audio signal and user historical data of an audio device are obtained, the audio signal including information of left and right channels, an audio signal and user historical data of an audio device are obtained, the audio signal including information of left and right channels, ambient noise in the audio signal is analyzed to obtain a first analysis result, voice information in the audio signal is analyzed to obtain a second analysis result, the type of music in the audio signal is analyzed to obtain a third analysis result, a weight analysis is performed based on the first analysis result, the second analysis result, and the third analysis result to obtain a current usage scenario, a frequency response analysis of the left and right channels is performed based on the audio signal, a phase analysis of the left and right channels is performed based on the audio signal, a time delay analysis of the left and right channels is performed based on the audio signal, and user habit analysis and scene adaptability analysis are performed according to the user historical data and the current usage scenario; parameters of the left and right channels of the audio device are corrected according to the channel characteristic analysis results, the user habit analysis results, and the scene adaptability analysis results. The solution conducts in-depth analysis of ambient noise, voice information, and music types in audio signals, combined with historical user data, to accurately identify current usage scenarios, and further conducts detailed analysis of the frequency response, phase, and time delay of the left and right channels of the audio device. Taking into account the user's personalized listening habits and scene adaptability, the solution intelligently corrects the channel parameters, significantly improving the accuracy of channel correction of audio equipment, and providing users with a listening experience that is more in line with personal preferences and environmental needs.

[0092] The above describes the vocal tract correction method in the embodiment of the present invention. The following describes the vocal tract correction device in the embodiment of the present invention. Figure 3 , an embodiment of the vocal tract correction device in the embodiment of the present invention includes:

[0093] An acquisition module 301 is used to acquire an audio signal output by an audio device and user history data, wherein the audio signal includes information of left and right channels;

[0094] An identification module 302, configured to identify a current usage scenario based on an audio signal;

[0095] A first analysis module 303, configured to analyze the channel characteristics of the left and right channels based on the audio signal;

[0096] The second analysis module 304 is used to perform user habit analysis and scenario adaptability analysis based on user historical data and current usage scenarios;

[0097] The correction module 305 is used to correct the parameters of the left and right channels of the audio device according to the channel characteristic analysis results, the user habit analysis results and the scene adaptability analysis results.

[0098] In an embodiment of the present invention, an audio signal and user history data of an audio device are obtained, the audio signal including information of two left and right channels; the current usage scenario is identified based on the audio signal; channel characteristics analysis of the left and right channels is performed based on the audio signal; user habit analysis and scene adaptability analysis are performed based on the user history data and the current usage scenario; and parameters of the left and right channels of the audio device are corrected based on the channel characteristics analysis results, the user habit analysis results, and the scene adaptability analysis results. This solution obtains audio signals and user history data, accurately identifies the current usage scenario, deeply analyzes the channel characteristics of the left and right channels, and comprehensively considers user habits and scene adaptability, thereby achieving refined correction of the left and right channel parameters of the audio device, significantly improving the accuracy and pertinence of the channel correction of the audio device.

[0099] See also Figure 4 Another embodiment of the vocal tract correction device in the embodiment of the present invention includes:

[0100] An acquisition module 301 is used to acquire an audio signal output by an audio device and user history data, wherein the audio signal includes information of left and right channels;

[0101] An identification module 302, configured to identify a current usage scenario based on an audio signal;

[0102] A first analysis module 303, configured to analyze the channel characteristics of the left and right channels based on the audio signal;

[0103] The second analysis module 304 is used to perform user habit analysis and scenario adaptability analysis based on user historical data and current usage scenarios;

[0104] The correction module 305 is used to correct the parameters of the left and right channels of the audio device according to the channel characteristic analysis results, the user habit analysis results and the scene adaptability analysis results.

[0105] Optionally, the identification module 302 includes:

[0106] The first analysis unit 3021 is used to analyze the environmental noise in the audio signal to obtain a first analysis result;

[0107] The second analysis unit 3022 is used to analyze the voice information in the audio signal to obtain a second analysis result;

[0108] The third analysis unit 3023 is used to analyze the music type in the audio signal to obtain a third analysis result;

[0109] The fourth analysis unit 3024 is used to perform weight analysis based on the first analysis result, the second analysis result and the third analysis result to obtain the current usage scenario.

[0110] Optionally, the first analysis unit 3021 may be specifically configured to:

[0111] Extracting noise characteristics of ambient noise in audio signals;

[0112] The noise feature is matched with a preset noise database; a first analysis result is obtained based on the matching result, and the first analysis result includes at least one usage scenario.

[0113] Optionally, the second analysis unit 3022 may be specifically configured to:

[0114] Perform speech recognition on the audio signal and convert the speech into text information; analyze the semantic content of the text information and determine the category or topic to which the speech belongs based on the semantic content; determine a second analysis result based on the category or topic, and the second analysis result includes at least one usage scenario.

[0115] Optionally, the third analysis unit 3023 may be specifically used for:

[0116] Extract music features from an audio signal; input the music features into a preset music classification model to obtain an audio type; determine the music style to which the music type belongs; and determine a third analysis result based on the music style, wherein the third analysis result includes at least one usage scenario.

[0117] Optionally, the fourth analyzing unit 3024 may be specifically configured to:

[0118] The total weight of each usage scenario is calculated based on the first analysis result, the second analysis result, and the third analysis result; and the usage scenario with the highest total weight is selected as the current usage scenario.

[0119] Optionally, the first analysis module 303 includes:

[0120] A frequency analysis unit 3031, used for performing frequency response analysis of left and right channels based on the audio signal;

[0121] A phase analysis unit 3032, configured to perform phase analysis of left and right channels based on an audio signal;

[0122] The time delay analysis unit 3033 is used to perform time delay analysis of left and right channels based on the audio signal.

[0123] Optionally, the frequency analysis unit 3031 may be specifically used for:

[0124] Obtain frequency response curves of the left and right channels based on the audio signal; determine the gain imbalance of the left and right channels within a specific frequency range using the difference in the frequency response curves; and analyze whether the frequency response of the left and right channels meets the preset consistency requirements based on the gain imbalance.

[0125] Optionally, the phase analysis unit 3032 may be specifically used for:

[0126] The phase spectra of the left and right channels are obtained based on the audio signal; the phase spectra of the left and right channels are compared to identify the spectrum area of ​​phase deviation or phase distortion; and the phase response of the left and right channels is analyzed to see whether it meets the preset phase requirement according to the degree of phase deviation or phase distortion.

[0127] Optionally, the time delay analysis unit 3033 may be specifically used for:

[0128] The time delay between the left and right channels is measured based on the audio signal; the synchronization index between the left and right channels is calculated based on the time delay; and whether the synchronization of the left and right channels meets the preset delay requirement is analyzed based on the synchronization index.

[0129] Optionally, the second analysis module 304 may be specifically used for:

[0130] Extract user behavior information from user historical data, including audio playback records, device settings, and user feedback, and analyze user usage preferences based on the behavior information; determine scenario preferences based on the current usage scenario.

[0131] Optionally, the correction module 305 may be specifically used for:

[0132] Prioritize and weight all parameters according to the degree of influence of preset audio parameters on audio playback; build a specific correction strategy for each parameter based on the results of channel characteristics analysis, user habit analysis and scene adaptability analysis; correct the parameters of the left and right channels of the audio device according to the correction strategy.

[0133] In an embodiment of the present invention, an audio signal and user historical data of an audio device are obtained, the audio signal including information of left and right channels, an audio signal and user historical data of an audio device are obtained, the audio signal including information of left and right channels, ambient noise in the audio signal is analyzed to obtain a first analysis result, voice information in the audio signal is analyzed to obtain a second analysis result, the type of music in the audio signal is analyzed to obtain a third analysis result, a weight analysis is performed based on the first analysis result, the second analysis result, and the third analysis result to obtain a current usage scenario, a frequency response analysis of the left and right channels is performed based on the audio signal, a phase analysis of the left and right channels is performed based on the audio signal, a time delay analysis of the left and right channels is performed based on the audio signal, and user habit analysis and scene adaptability analysis are performed according to the user historical data and the current usage scenario; parameters of the left and right channels of the audio device are corrected according to the channel characteristic analysis results, the user habit analysis results, and the scene adaptability analysis results. The solution conducts in-depth analysis of ambient noise, voice information, and music types in audio signals, combined with historical user data, to accurately identify current usage scenarios, and further conducts detailed analysis of the frequency response, phase, and time delay of the left and right channels of the audio device. Taking into account the user's personalized listening habits and scene adaptability, the solution intelligently corrects the channel parameters, significantly improving the accuracy of channel correction of audio equipment, and providing users with a listening experience that is more in line with personal preferences and environmental needs.

[0134] above Figure 3 and Figure 4 The vocal channel correction device in the embodiment of the present invention is described in detail from the perspective of modular functional entities. The vocal channel correction device in the embodiment of the present invention is described in detail from the perspective of hardware processing.

[0135] See also Figure 5 As shown, the vocal tract correction device includes a processor 500 and a memory 501 . The memory 501 stores machine executable instructions that can be executed by the processor 500 . The processor 500 executes the machine executable instructions to implement the above-mentioned vocal tract correction method.

[0136] Further, Figure 5 The acoustic channel correction device shown further includes a bus 502 and a communication interface 503 , and the processor 500 , the communication interface 503 and the memory 501 are connected via the bus 502 .

[0137] Among them, the memory 501 may include a high-speed random access memory (RAM), and may also include a non-volatile memory (non-volatile memory), for example, at least one disk storage. The communication connection between the system network element and at least one other network element is realized through at least one communication interface 503 (which can be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. can be used. The bus 502 can be an ISA bus, a PCI bus or an EISA bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 5 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0138] The processor 500 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit or software instructions in the processor 500. The above processor 500 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The disclosed methods, steps and logic block diagrams in the embodiments of the present disclosure can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in conjunction with the embodiments of the present disclosure can be directly embodied as a hardware decoding processor to be executed, or a combination of hardware and software modules in the decoding processor can be executed. The software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 501 , and the processor 500 reads the information in the memory 501 and completes the method steps of the above-mentioned embodiment in combination with its hardware.

[0139] The present invention also provides a vocal tract correction device, wherein the computer device comprises a memory and a processor, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the processor executes the steps of the vocal tract correction method in the above-mentioned embodiments.

[0140] The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium. When the instructions are executed on a computer, the computer executes the steps of the vocal tract correction method.

[0141] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0142] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program codes.

[0143] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A vocal tract correction method, characterized in that: The vocal tract correction method comprises: Acquire an audio signal of an audio device and user history data, wherein the audio signal includes information of two left and right channels; identifying a current usage scenario based on the audio signal; Performing channel characteristic analysis of left and right channels based on the audio signal; Performing user habit analysis and scenario adaptability analysis based on the user historical data and the current usage scenario; The parameters of the left and right channels of the audio device are corrected according to the channel characteristic analysis results, the user habit analysis results and the scene adaptability analysis results.

2. The vocal tract correction method according to claim 1, characterized in that: The identifying the current usage scenario based on the audio signal includes: Analyze the environmental noise in the audio signal to obtain a first analysis result; Analyzing the voice information in the audio signal to obtain a second analysis result; Analyze the music type in the audio signal to obtain a third analysis result; A weight analysis is performed based on the first analysis result, the second analysis result, and the third analysis result to obtain a current usage scenario.

3. The vocal tract correction method according to claim 2, characterized in that: The step of analyzing the ambient noise in the audio signal to obtain a first analysis result includes: Extracting noise characteristics of ambient noise in the audio signal; Matching the noise characteristics with a preset noise database; A first analysis result is obtained based on the matching result, where the first analysis result includes at least one usage scenario.

4. The vocal tract correction method according to claim 2, characterized in that: The analyzing the voice information in the audio signal to obtain a second analysis result includes: Performing speech recognition on the audio signal and converting the speech into text information; Analyzing the semantic content of the text information, and determining the category or topic to which the speech belongs based on the semantic content; A second analysis result is determined according to the category or theme, wherein the second analysis result includes at least one usage scenario.

5. The vocal tract correction method according to claim 2, characterized in that: The analyzing the music type in the audio signal to obtain a third analysis result includes: Extracting music features from the audio signal; Inputting the music features into a preset music classification model to obtain an audio type; Determine the music style to which the music type belongs; A third analysis result is determined according to the music style, and the third analysis result includes at least one usage scenario.

6. The vocal tract correction method according to claim 2, characterized in that: The performing weight analysis based on the first analysis result, the second analysis result, and the third analysis result to obtain the current usage scenario includes: Calculate the total weight of each usage scenario based on the first analysis result, the second analysis result, and the third analysis result; The usage scenario with the highest total weight is selected as the current usage scenario.

7. The vocal tract correction method according to claim 1, characterized in that: The performing channel characteristic analysis of the left and right channels based on the audio signal comprises: Performing frequency response analysis of left and right channels based on the audio signal; Performing phase analysis of left and right channels based on the audio signal; A time delay analysis of the left and right channels is performed based on the audio signal.

8. The vocal tract correction method according to claim 7, characterized in that: The frequency response analysis of the left and right channels based on the audio signal comprises: Acquire frequency response curves of left and right channels based on the audio signal; Determining the gain imbalance of the left and right channels within a specific frequency range using the difference in the frequency response curves; The frequency responses of the left and right channels are analyzed according to the gain imbalance to determine whether they meet a preset consistency requirement.

9. The vocal tract correction method according to claim 7, characterized in that: Performing phase analysis of left and right channels based on the audio signal includes: Acquire phase spectra of left and right channels based on the audio signal; Comparing the phase spectra of the left and right channels to identify the spectral regions where phase deviation or phase distortion occurs; The phase response of the left and right channels is analyzed according to the degree of the phase deviation or phase distortion to determine whether it meets the preset phase requirement.

10. The vocal tract correction method according to claim 7, characterized in that: The time delay analysis of the left and right channels based on the audio signal comprises: Measuring a time delay between left and right channels based on the audio signal; Calculating a synchronization index between the left and right channels based on the time delay; It is analyzed according to the synchronization index whether the synchronization of the left and right channels meets the preset delay requirement.

11. The vocal tract correction method according to claim 1, characterized in that: The performing of user habit analysis and scenario adaptability analysis based on the user historical data and the current usage scenario includes: Extracting user behavior information from the user history data, the behavior information including audio playback records, device settings and user feedback, and analyzing the user's usage preferences based on the behavior information; A scenario preference is determined according to the current usage scenario.

12. The vocal tract correction method according to claim 1, characterized in that: The correcting the parameters of the left and right channels of the audio device according to the channel characteristic analysis result, the user habit analysis result and the scene adaptability analysis result includes: Determining parameters that need to be corrected according to the vocal tract analysis results; Allocating priorities and weights to all parameters based on the parameter characteristics of the parameters that need to be corrected and the degree of influence on audio playback; Prioritize and weight all parameters according to the influence of preset audio parameters on audio playback; Based on the vocal tract characteristic analysis result, the user habit analysis result and the scene adaptability analysis result, constructing a specific correction strategy for each parameter; The parameters of the left and right channels of the audio device are corrected according to the correction strategy.

13. A vocal tract correction device, characterized in that: The vocal tract correction device comprises: An acquisition module, used to acquire an audio signal output by an audio device and user history data, wherein the audio signal includes information of left and right channels; An identification module, configured to identify a current usage scenario based on the audio signal; A first analysis module, configured to perform channel characteristic analysis of left and right channels based on the audio signal; A second analysis module, used to perform user habit analysis and scenario adaptability analysis based on the user historical data and the current usage scenario; The correction module is used to correct the parameters of the left and right channels of the audio device according to the channel characteristic analysis results, the user habit analysis results and the scene adaptability analysis results.

14. A vocal tract correction device, characterized in that: The vocal tract correction device comprises: a memory and at least one processor, wherein instructions are stored in the memory; The at least one processor calls the instructions in the memory to enable the vocal channel correction device to perform the vocal channel correction method as described in any one of claims 1-12.

15. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions are executed by a processor, the vocal tract correction method according to any one of claims 1 to 12 is implemented.

Images