Audio equipment mode language control method and system

Abstract

The invention relates to the technical field of voice control, and discloses an audio equipment mode language control method, which comprises the following steps: acquiring a voice instruction, extracting an instruction audio, and performing energy enhancement on the audio to obtain an enhanced audio; analyzing a power spectrum of the enhanced audio, and performing noise suppression on the enhanced audio based on the power spectrum to obtain a suppressed audio; determining a time-frequency adjustment audio of the suppressed audio, querying an optimal codebook of the time-frequency adjustment audio, and identifying audio features of the suppressed audio; analyzing an instruction text corresponding to the voice instruction, and performing mode switching on the audio equipment based on the instruction text to obtain target audio equipment; analyzing the audio effect of the target audio equipment; if the audio effect does not meet the preset user demand effect, performing parameter adjustment on the equipment; and if the audio effect meets the preset user demand effect, taking the target audio equipment as audio control equipment of the voice instruction. The accuracy of audio equipment control can be improved.

Description

Technical Field

[0001] The present invention relates to the field of voice control technology, and in particular to a method and system for controlling a mode language of an audio device. Background Art

[0002] Audio devices play an important role in people's daily life and work. Building a mode language control platform for audio devices allows users to easily switch between different modes of audio devices through voice commands, thereby improving the convenience and efficiency of using audio devices. At the same time, users can quickly adjust the parameter settings of audio devices according to different scene requirements to obtain a more personalized audio experience, and facilitate the interaction and collaboration between different users and between audio devices and other smart devices, thereby reducing the trouble caused by operational complexity and improving user satisfaction.

[0003] At present, the control of audio equipment usually adopts traditional control methods such as manual buttons, touch screen operations or remote controls. However, with this traditional control method, users need to perform cumbersome operations, and it is difficult to achieve precise control in some cases (such as environmental noise interference, fuzzy or unclear voice, etc.). In addition, due to the diverse usage scenarios of audio equipment, there are a large number of different modes and parameter setting requirements. When faced with complex operations and a large number of instructions, the traditional control method is not efficient and accurate in processing data, which makes the control of audio equipment not convenient enough. Summary of the invention

[0004] The present invention provides an audio device mode language control method and system, the main purpose of which is to improve the accuracy of audio device control.

[0005] To achieve the above object, the present invention provides an audio device mode language control method, comprising:

[0006] Acquire a voice command of a user corresponding to the audio device, extract voice audio of the voice command, and perform energy enhancement on the voice audio to obtain enhanced audio;

[0007] Analyze the power spectrum of the enhanced audio, perform signal filtering on the enhanced audio based on the power spectrum to obtain filtered audio, calculate the long-time frame power of the filtered audio, and perform noise suppression on the filtered audio based on the long-time frame power to obtain suppressed audio;

[0008] Querying the multi-channel time-frequency and power of the suppressed audio, performing time-frequency scale adjustment on the channel time-frequency and the power to obtain an adjusted time-frequency and an adjusted power, determining the time-frequency adjusted audio of the suppressed audio based on the adjusted time-frequency and the adjusted power, querying the optimal codebook of the time-frequency adjusted audio, and identifying the audio features of the suppressed audio based on the optimal codebook;

[0009] Based on the audio feature, parsing the instruction text corresponding to the voice instruction, and based on the instruction text, switching the mode of the audio device to obtain a target audio device;

[0010] Analyzing the audio effect of the target audio device;

[0011] If the audio effect does not meet the preset user demand effect, query the user feedback instruction of the target audio device, adjust the parameters of the target audio device based on the user feedback instruction, and then return to the step of analyzing the audio effect of the target audio device;

[0012] If the audio effect meets the preset user requirement, the target audio device is used as the audio control device of the voice command.

[0013] Optionally, the performing energy enhancement on the speech audio to obtain enhanced audio includes:

[0014] Performing frequency domain conversion on the speech audio to obtain converted audio;

[0015] Calculating energy levels of different frequency intervals in the converted audio to obtain regional audio energy levels;

[0016] Determining an energy enhancement interval of the converted audio based on the regional audio energy level;

[0017] The converted audio is energy enhanced based on the energy enhancement interval to obtain enhanced audio.

[0018] Optionally, analyzing the power spectrum of the enhanced audio includes:

[0019] Extracting an initial signal of the enhanced audio;

[0020] Calculate the wavelet coefficient spectrum of the initial signal;

[0021] Based on the wavelet coefficient spectrum, a power spectrum of the enhanced audio is calculated.

[0022] Optionally, the performing signal filtering on the enhanced audio based on the power spectrum to obtain filtered audio includes:

[0023] Based on the power spectrum, determining a filtering range of the enhanced audio;

[0024] Based on the filtering range, calculating the signal pass rate of the enhanced audio;

[0025] Based on the signal pass rate, the signal of the enhanced audio is transformed to obtain filtered audio.

[0026] Optionally, calculating the long-time frame power of the filtered audio includes:

[0027] Performing audio segmentation on the filtered audio to obtain segmented audio;

[0028] Calculating the signal power of the split audio;

[0029] Based on the frame power, the long-term frame power of the filtered audio is calculated.

[0030] Optionally, the querying of the optimal codebook for the time-frequency adjusted audio includes:

[0031] Converting the time-frequency adjusted audio into an audio vector;

[0032] Performing vector clustering on the audio vector to obtain a clustering vector;

[0033] Querying the cluster center of the cluster vector;

[0034] Based on the cluster center, clustering optimization is performed on the cluster vector to obtain an optimized cluster vector;

[0035] Screening the optimized clustering vector to obtain a screened clustering vector;

[0036] Based on the screened clustering vectors, an optimal codebook for the time-frequency adjusted audio is determined.

[0037] Optionally, converting the time-frequency adjusted audio into an audio vector includes:

[0038] Performing frame processing on the time-frequency adjusted audio to obtain multiple frames of audio;

[0039] Performing frequency scale conversion on the multiple frames of audio to obtain converted audio;

[0040] Perform discrete cosine transform on the converted audio to obtain an audio vector.

[0041] Optionally, the identifying audio features of the suppressed audio based on the optimal codebook includes:

[0042] Querying characteristic parameters of the audio suppression;

[0043] Performing vectorization processing on the characteristic parameters to obtain a parameter vector;

[0044] Matching the parameter vector with the optimal codebook to obtain a matching result;

[0045] Based on the matching result, the suppressed audio is marked with time-frequency domain features to determine audio features of the suppressed audio.

[0046] Optionally, analyzing the audio effect of the target audio device includes:

[0047] Querying control parameters corresponding to the target audio device;

[0048] Performing an audio effect test on the target audio device according to the control parameters;

[0049] After the audio effect test is completed, performing intelligent voice interaction with a user corresponding to the target audio device;

[0050] Based on the interaction result of the intelligent voice interaction, an audio effect of the target audio device is determined.

[0051] In order to solve the above problems, the present invention also provides an audio device mode language control system, the system comprising:

[0052] The audio enhancement module is used to obtain the voice command of the user corresponding to the audio device, extract the voice audio of the voice command, and perform energy enhancement on the voice audio to obtain enhanced audio;

[0053] an audio suppression module, configured to analyze a power spectrum of the enhanced audio, perform signal filtering on the enhanced audio based on the power spectrum to obtain filtered audio, calculate a long-time frame power of the filtered audio, and perform noise suppression on the filtered audio based on the long-time frame power to obtain suppressed audio;

[0054] An audio feature recognition module is used to query the multi-channel time-frequency and power of the suppressed audio, perform time-frequency scale adjustment on the channel time-frequency and the power to obtain an adjusted time-frequency and an adjusted power, determine the time-frequency adjusted audio of the suppressed audio based on the adjusted time-frequency and the adjusted power, query an optimal codebook for the time-frequency adjusted audio, and identify audio features of the suppressed audio based on the optimal codebook;

[0055] A device mode switching module, used to parse the instruction text corresponding to the voice instruction based on the audio feature, and switch the mode of the audio device based on the instruction text to obtain a target audio device;

[0056] An audio effect analysis module, used to analyze the audio effect of the target audio device;

[0057] a device adjustment module, configured to query a user feedback instruction of the target audio device if the audio effect does not meet a preset user requirement effect, adjust parameters of the target audio device based on the user feedback instruction, and then return to the step of analyzing the audio effect of the target audio device;

[0058] The target device acquisition module is used to use the target audio device as the audio control device of the voice command if the audio effect meets the preset user requirement effect.

[0059] The present invention can enhance the energy of the voice audio to obtain enhanced audio, which can make the voice signal more prominent and improve the accuracy and reliability of subsequent processing; the present invention can understand the energy distribution of the audio signal at different frequencies by analyzing the power spectrum of the enhanced audio, so as to better remove audio noise; querying the multi-channel time-frequency and power of the suppressed audio can more accurately capture the changing characteristics of the audio signal of the suppressed audio in time and frequency; the present invention can clarify the instructions issued by the user by parsing the instruction text corresponding to the voice instruction based on the audio features, so as to accurately control the audio device. Therefore, the present invention can improve the accuracy of audio device control. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 A flowchart of a method for controlling a mode language of an audio device provided by an embodiment of the present invention;

[0061] Figure 2 A functional module diagram of an audio device mode language control system provided by an embodiment of the present invention;

[0062] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0063] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0064] The embodiment of the present application provides a method for controlling the mode language of an audio device. The execution subject of the method for controlling the mode language of an audio device includes but is not limited to at least one of the electronic devices such as a server, a terminal, etc. that can be configured to execute the method provided by the embodiment of the present application. In other words, the method for controlling the mode language of an audio device can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc. The server can be an independent server, or it can be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), and big data and artificial intelligence platforms.

[0065] Reference Figure 1FIG. 1 is a flow chart of a method for controlling a language in an audio device mode provided by an embodiment of the present invention. In this embodiment, the method for controlling a language in an audio device mode includes:

[0066] S1. Acquire a voice command of a user corresponding to an audio device, extract voice audio of the voice command, and perform energy enhancement on the voice audio to obtain enhanced audio.

[0067] The embodiment of the present invention can obtain the reference of the operation of the audio device by obtaining the voice command of the user corresponding to the audio device. The audio device refers to a hardware device that can receive, process and output audio signals, and the voice command refers to a specific command sentence expressed by voice, and the audio device performs the corresponding operation by recognizing these commands. The voice command can be obtained by receiving the user's voice signal through an audio input device (such as a microphone) in the audio device.

[0068] Furthermore, the embodiment of the present invention can allow the device to focus only on the user's voice signal by extracting the voice audio of the voice command, eliminating other irrelevant background noise interference.

[0069] The voice audio refers to a voice signal, which can be obtained by separating the user's voice part from the received audio signal through filtering technology.

[0070] Furthermore, the embodiment of the present invention can enhance the energy of the speech audio to obtain enhanced audio, which can make the speech signal more prominent and improve the accuracy and reliability of subsequent processing.

[0071] As an embodiment of the present invention, the energy enhancement of the speech audio to obtain enhanced audio includes: performing frequency domain conversion on the speech audio to obtain converted audio, calculating energy levels of different frequency intervals in the converted audio to obtain regional audio energy levels, determining an energy enhancement interval of the converted audio based on the regional audio energy levels, and performing energy enhancement on the converted audio based on the energy enhancement interval to obtain enhanced audio.

[0072] Optionally, the converted audio can be obtained by performing frequency domain conversion on the speech audio through fast Fourier transform, the energy levels of the different frequency intervals can be determined by calculating the sum of squares of the audio signals corresponding to the speech audio, the energy enhancement interval can be obtained by comparing the energy of different intervals with a preset energy threshold, if the energy of the different intervals is less than the energy threshold, enhancement is required, the preset energy threshold can be set to 0.5J, or a suitable value can be set according to the actual application scenario, the energy enhancement of the converted audio to obtain the enhanced audio can be performed by a spectrum enhancement method, that is, processing the audio signal in the frequency domain to enhance the energy of a specific frequency interval, such as enhancing the energy of the main frequency components in the speech signal.

[0073] S2. Analyze the power spectrum of the enhanced audio, perform signal filtering on the enhanced audio based on the power spectrum to obtain filtered audio, calculate the long-time frame power of the filtered audio, and perform noise suppression on the filtered audio based on the long-time frame power to obtain suppressed audio.

[0074] The embodiment of the present invention can understand the energy distribution of the audio signal at different frequencies by analyzing the power spectrum of the enhanced audio, so as to better remove the audio noise. Wherein, the power spectrum refers to a representation method used to describe the distribution of signal power at different frequencies.

[0075] As an embodiment of the present invention, the analyzing the power spectrum of the enhanced audio includes: extracting the initial signal of the enhanced audio, calculating the wavelet coefficient spectrum of the initial signal, and calculating the power spectrum of the enhanced audio based on the wavelet coefficient spectrum using the following formula:

[0076]

[0077] Wherein, P(ω) represents the power spectrum, ω represents the wavelet coefficient spectrum, E represents the time-frequency conversion value, and T represents the total number of frames of the enhanced audio.

[0078] Optionally, the wavelet coefficient spectrum of the initial signal can be obtained by calculating a wavelet function.

[0079] Furthermore, the embodiment of the present invention performs signal filtering on the enhanced audio based on the power spectrum to obtain filtered audio, which can remove unnecessary frequency components and reduce the amount of audio data, thereby reducing the computational complexity and storage requirements of subsequent processing.

[0080] As an embodiment of the present invention, the signal filtering of the enhanced audio based on the power spectrum to obtain the filtered audio includes: determining a filtering range of the enhanced audio based on the power spectrum, and calculating the signal pass rate of the enhanced audio based on the filtering range using the following formula:

[0081]

[0082] Among them, H(f) represents the signal pass, f represents a frequency in the enhanced audio, and f l Indicates the minimum value of the filtering range, f h Indicates the maximum value of the filtering range;

[0083] Based on the signal pass rate, the signal of the enhanced audio is transformed to obtain filtered audio.

[0084] The signal pass rate refers to the degree of passage of the input signal at different frequency components in the filter. When H(f)=1, it means that the signal at frequency f completely passes through the filter; when H(f)=0, it means that the signal at this frequency is completely blocked.

[0085] Optionally, the filtering range of the enhanced audio based on the power spectrum can be determined by analyzing the peaks and valleys of the power spectrum. For example, if the power spectrum has an obvious peak at a certain frequency, it may mean that the frequency corresponds to the fundamental frequency of the signal or an important harmonic component. For periodic signals, the peaks in the power spectrum will appear at integer multiples of the periodic frequency. The periodic characteristics of the signal can be determined based on these peaks, and the relevant frequency components can be retained. The signal of the enhanced audio is transformed based on the signal pass-through to obtain the filtered audio. The signal of the enhanced audio corresponding to the signal pass-through is Fourier transformed to obtain a filtered transformed signal, and then the filtered audio is obtained by inverse Fourier transforming the filtered transformed signal.

[0086] The embodiment of the present invention can help users analyze the noise characteristics in the audio by calculating the long-time frame power of the filtered audio. By observing the changes in the long-time frame power in different time periods, it can be determined whether the intensity of the noise is stable and the frequency distribution of the noise.

[0087] The long-term frame power refers to a comprehensive index obtained by statistically analyzing the audio signal power of multiple consecutive frames.

[0088] As an embodiment of the present invention, the calculating the long-time frame power of the filtered audio includes: performing audio segmentation on the filtered audio to obtain segmented audio, and calculating the signal power of the segmented audio using the following formula:

[0089]

[0090] Among them, P framerepresents the signal power, N represents the frame length of the segmented audio, w(n) represents the weighting function, and x(n) represents the value of the audio signal corresponding to the segmented audio at the nth sampling point;

[0091] Based on the frame power, the long-time frame power of the filtered audio is calculated using the following formula:

[0092]

[0093] Among them, P long-term represents the long-term frame power, M represents the total number of frames of filtered audio, and P frame (i) represents the i-th frame of filtered audio.

[0094] Furthermore, the embodiment of the present invention suppresses noise on the filtered audio based on the long-time frame power to obtain suppressed audio, which can improve signal processing efficiency and further optimize audio processing system performance.

[0095] Optionally, the suppressed audio may be noise suppressed by utilizing the long-time frame power in combination with an asymmetric filter function.

[0096] S3. Query the multi-channel time-frequency and power of the suppressed audio, perform time-frequency scale adjustment on the channel time-frequency and the power to obtain adjusted time-frequency and adjusted power, determine the time-frequency adjusted audio of the suppressed audio based on the adjusted time-frequency and adjusted power, query the optimal codebook of the time-frequency adjusted audio, and identify the audio features of the suppressed audio based on the optimal codebook.

[0097] The embodiment of the present invention can more accurately capture the variation characteristics of the audio signal of the suppressed audio in time and frequency by querying the multi-channel time-frequency and power of the suppressed audio.

[0098] The multi-channel time-frequency refers to the changes of the audio signal over time and frequency on multiple channels, and the power refers to the energy intensity of the audio signal in a specific time period.

[0099] Furthermore, the embodiment of the present invention can make the characteristics of the audio signal more obvious by adjusting the time-frequency scale of the channel time-frequency and the power to obtain the adjusted time-frequency and adjusted power, thereby facilitating feature extraction and analysis.

[0100] Optionally, the time-frequency scale of the channel time-frequency and the power is adjusted, and the adjusted time-frequency and adjusted power can be adjusted by a resampling method, and the resampling parameters are determined according to the time-frequency scale to be adjusted. The resampling parameters include the change ratio of the sampling rate, the filter type, etc.

[0101] Furthermore, the embodiment of the present invention can quickly find the codeword that best matches the data by querying the optimal codebook of the time-frequency adjusted audio, thereby achieving effective processing and classification of the data.

[0102] The codebook refers to a set used to encode or represent data. In audio processing, a codebook is usually composed of some specific codewords, which can be used to represent the characteristics or patterns of audio.

[0103] As an embodiment of the present invention, the querying of the optimal codebook of the time-frequency adjusted audio includes: converting the time-frequency adjusted audio into an audio vector, performing vector clustering on the audio vector to obtain a cluster vector, querying the cluster center of the cluster vector, performing cluster optimization on the cluster vector based on the cluster center to obtain an optimized cluster vector, screening the optimized cluster vector to obtain a screened cluster vector, and determining the optimal codebook of the time-frequency adjusted audio based on the screened cluster vector.

[0104] Optionally, the vector clustering of the audio vector may be implemented by a K-Means method, such as first randomly selecting K initial cluster centers, then calculating the distance between each audio vector and each cluster center, assigning the audio vector to the cluster where the nearest cluster center is located, and then recalculating the center of each cluster, and repeating this process until the cluster center no longer changes or a certain number of iterations is reached; it should be noted that after completing the vector clustering, each cluster has a center, which can be obtained by calculating the mean value of all audio vectors in the cluster. For example, when using K-Means clustering, the center of each cluster is the center of the last iteration. The K points determined in the generation are clustered based on the cluster center, and the cluster vector is clustered and optimized to obtain the optimized cluster vector. The initial clustering result can be further refined by fuzzy C-means clustering (FCM); the optimized cluster vector is screened to obtain the screened cluster vector by calculating the distance from each audio vector to the center of the cluster to which it belongs. If the distance exceeds a preset threshold, the audio vector is considered to be an outlier and is removed. It should be noted that the preset threshold needs to be set in combination with the actual application operation. The optimal codebook can be obtained by taking the center of the screened cluster vector as the codeword of the codebook, and these codewords constitute the optimal codebook.

[0105] Furthermore, the embodiment of the present invention can accurately classify different types of suppressed audio by identifying the audio features of the suppressed audio based on the optimal codebook, for example, classifying the audio into different categories such as speech, music, and ambient sound, and the speech recognition system can identify the identity of the speaker, language content, and other information. For example, in an intelligent voice assistant application, the user's voice commands can be accurately identified to better provide services to the user.

[0106] As an optional embodiment of the present invention, converting the time-frequency adjusted audio into an audio vector includes: performing frame processing on the time-frequency adjusted audio to obtain multiple frames of audio, performing frequency scale conversion on the multiple frames of audio to obtain converted audio, and performing discrete cosine transform on the converted audio to obtain an audio vector.

[0107] Optionally, the frame processing of the time-frequency adjusted audio to obtain multiple frames of audio may be implemented through a window function, and the frequency scale conversion of the multiple frames of audio to obtain converted audio may be implemented through Mel frequency scale conversion.

[0108] As an embodiment of the present invention, the method of identifying the audio features of the suppressed audio based on the optimal codebook includes: querying the characteristic parameters of the suppressed audio, vectorizing the characteristic parameters to obtain a parameter vector, matching the parameter vector with the optimal codebook to obtain a matching result, and based on the matching result, marking the suppressed audio in the time-frequency domain to determine the audio features of the suppressed audio. The characteristic parameters refer to numerical values ​​used to describe specific attributes of audio signals. These parameters can reflect the characteristics of the audio from different angles and provide a basis for tasks such as audio analysis, recognition, and classification.

[0109] Optionally, the characteristic parameters can be obtained through Mel frequency cepstral coefficients; the vectorization of the characteristic parameters to obtain the parameter vector can be converted through a vector matrix; the matching of the parameter vector with the optimal code book to obtain the matching result can be performed by using a distance measurement method to match the parameter vector with the optimal code book, such as Euclidean distance, cosine distance and other measurement methods; based on the matching result, the suppressed audio is marked with time-frequency domain features to determine the audio features of the suppressed audio. The time-frequency domain features of the suppressed audio can be determined based on the matched codewords. It should be explained that each codeword represents a specific audio feature, and each codeword in the codebook can be given a specific time-frequency domain feature label in advance. For example, a codeword in the codebook represents a high-frequency speech feature. Then, when the parameter vector matches this codeword, the high-frequency speech feature label can be marked for the suppressed audio. By matching and marking different parts of the entire suppressed audio, the overall audio features of the suppressed audio can be determined.

[0110] S4. Based on the audio feature, parse the instruction text corresponding to the voice instruction, and based on the instruction text, switch the mode of the audio device to obtain a target audio device.

[0111] The embodiment of the present invention can clarify the instruction issued by the user by parsing the instruction text corresponding to the voice instruction based on the audio feature, so as to accurately control the audio device.

[0112] Optionally, the instruction text can be obtained by performing feature recognition and translation on the audio features using a convolutional neural network.

[0113] Furthermore, the embodiment of the present invention can improve the accuracy and reliability of instruction execution by switching the mode of the audio device based on the instruction text to obtain the target audio device, thereby ensuring that the audio device switches the mode according to the user's true intention.

[0114] As an embodiment of the present invention, the method of switching the mode of the audio device based on the instruction text to obtain the target audio device includes: performing natural language processing on the instruction text to obtain a processed text, performing intent recognition on the instruction text based on the processed text, and calculating the intent score of the intent recognition using the following formula:

[0115]

[0116] Among them, Score M (T) represents the intention score, w represents the keyword of the instruction text, T represents the instruction text, M represents a specific mode of the audio device, K M represents the keyword set of mode M, and Weight(w) represents the weight of w;

[0117] The audio device is switched in mode based on the intention score to obtain a target audio device. Optionally, the audio device is switched in mode based on the intention score to obtain a target audio device that can be switched according to a specific score value of the intention score. For example, there is a smart speaker device that supports Bluetooth mode, AUX mode and Wi-Fi music playback mode. For the Bluetooth mode, the keyword set K 蓝牙 ={"Bluetooth", "Connect Bluetooth", "Pair Bluetooth", "Switch to Bluetooth"}, the weights can be set as follows: "Bluetooth" weight is 0.5, "Connect Bluetooth" weight is 0.8, "Pair Bluetooth" weight is 0.7, "Switch to Bluetooth" weight is 0.9, "AUX" weight is 0.6, "Connect AUX" weight is 0.8, "Switch to AUX" weight is 0.85. For Wi-Fi music playback mode, the keyword set K Wi-Fi ={"W i-Fi", "Play music", "Connect to Wi-Fi", "Switch to Wi-Fi music"}, the weights are set as follows: "Wi-Fi" weight is 0.4, "Play music" weight is 0.6, "Connect to Wi-Fi" weight is 0.7, "Switch to Wi-Fi music" weight is 0.8, assuming that the user issues the command "Switch to Bluetooth mode", this command text T1 = "Switch to Bluetooth mode". One of the keywords in the keyword set, and the weight is 0.9), the intention score for AUX mode: Score AUX (T1) = 0 (the command text does not contain the keyword of AUX mode), the intention score for Wi-Fi music playback mode: Score Wi-Fi (T1) = 0 (the command text does not contain the keyword of Wi-Fi music playback mode). Assume that the user issues the command "I want to listen to music and play it through Wi-Fi connection", and the command text T2 = "I want to listen to music and play it through Wi-Fi connection". The intention score for Bluetooth mode is: Score 蓝牙 (T2) = 0, intention score for AUX mode: Score AUX (T2) = 0, the intention score for Wi-Fi music playback mode: Score Wi-Fi (T2) = 0.4 + 0.6 + 0.7 + 0.8 = 2.5 (the command text contains the keywords "Wi-Fi", "play music", "connect to Wi-Fi", and "switch to Wi-Fi music", and the corresponding weights are added respectively). Based on the calculated intention score, the smart speaker makes a mode switching decision.

[0118] For instruction T1, since the intent score of Bluetooth mode is the highest (0.9), while the intent scores of AUX mode and Wi-Fi music playback mode are both 0, the smart speaker switches to Bluetooth mode. For instruction T2, the intent score of Wi-Fi music playback mode is the highest (2.5), while the intent scores of Bluetooth mode and AUX mode are both 0, so the smart speaker switches to Wi-Fi music playback mode and starts playing music through the Wi-Fi connection.

[0119] S5. Analyze the audio effect of the target audio device.

[0120] The embodiment of the present invention can understand the accuracy of the audio device in executing the user's instructions by analyzing the audio effect of the target audio device.

[0121] As an embodiment of the present invention, the analyzing the audio effect of the target audio device includes: querying control parameters corresponding to the target audio device, performing an audio effect test on the target audio device according to the control parameters, and after the audio effect test is completed, performing intelligent voice interaction with a user corresponding to the target audio device, and determining the audio effect of the target audio device based on the interaction result of the intelligent voice interaction.

[0122] The control parameters refer to a series of specific values ​​or setting options used to adjust and control the performance of the target audio device and the audio output effect, such as volume-related parameters, sound effect mode parameters, and audio input and output parameters.

[0123] Optionally, the control parameters can be obtained by querying the user instructions in the instruction text, and the audio effect test of the target audio device based on the control parameters can make corresponding settings adjustments to the target audio device according to the control parameters, and the test is performed after the settings are adjusted. The user corresponding to the target audio device performs intelligent voice interaction through the voice interaction system of the audio device, and the audio effect of the target audio device based on the interaction result of the intelligent voice interaction is determined by identifying the user's feedback information. If the user indicates "ok" or "ok", it means that the audio effect is fine. If the user indicates "low volume" or "change channel mode", it means that there is a problem with the audio effect.

[0124] S6. If the audio effect does not meet the preset user demand effect, query the user feedback instruction of the target audio device, adjust the parameters of the target audio device based on the user feedback instruction, and then return to the step of analyzing the audio effect of the target audio device.

[0125] In the embodiment of the present invention, if the audio effect does not meet the preset user requirement effect, the user feedback instruction of the target audio device can be queried so that the setting of the audio device does not meet the user requirement and can be adjusted immediately to achieve the user's satisfaction. The user feedback instruction can be obtained by using the voice interaction system of the audio device to interact with the user.

[0126] It should be noted that if the audio effect does not meet the preset user demand effect, it means that the audio effect of the audio device does not meet the user's requirements, such as the settings of the volume, channel and playback mode do not meet the user's requirements.

[0127] Furthermore, the embodiment of the present invention can gradually adjust the control parameters of the audio device by returning to the step of analyzing the audio effect of the target audio device after adjusting the parameters of the target audio device based on the user feedback instruction until the user's control requirements for the audio device are met.

[0128] S7: If the audio effect meets the preset user requirement, use the target audio device as the audio control device of the voice command.

[0129] It should be noted that if the audio effect meets the preset user demand effect, it means that the audio effect of the audio device meets the user's requirements, such as the volume, channel and playback mode all meet the user's requirements, then the audio control device can be used to perform the user's audio control task.

[0130] like Figure 2 , which is a functional module diagram of an audio device mode language control system provided by an embodiment of the present invention.

[0131] The audio device mode language control system 200 of the present invention can be installed in an electronic device. According to the functions to be implemented, the audio device mode language control system 200 may include a data acquisition module 201, a data class center identification module 202, a data screening module 203 and a data storage module 204. The module of the present invention can also be called a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.

[0132] In this embodiment, the functions of each module / unit are as follows:

[0133] The audio enhancement module 201 is used to obtain a voice command of a user corresponding to the audio device, extract the voice audio of the voice command, and perform energy enhancement on the voice audio to obtain enhanced audio;

[0134] The audio suppression module 202 is used to analyze the power spectrum of the enhanced audio, perform signal filtering on the enhanced audio based on the power spectrum to obtain filtered audio, calculate the long-time frame power of the filtered audio, and perform noise suppression on the filtered audio based on the long-time frame power to obtain suppressed audio;

[0135] The audio feature recognition module 203 is used to query the multi-channel time-frequency and power of the suppressed audio, perform time-frequency scale adjustment on the channel time-frequency and the power to obtain an adjusted time-frequency and an adjusted power, determine the time-frequency adjusted audio of the suppressed audio based on the adjusted time-frequency and the adjusted power, query the optimal codebook of the time-frequency adjusted audio, and identify the audio features of the suppressed audio based on the optimal codebook;

[0136] The device mode switching module 204 is used to parse the instruction text corresponding to the voice instruction based on the audio feature, and switch the mode of the audio device based on the instruction text to obtain a target audio device;

[0137] The audio effect analysis module 205 is used to analyze the audio effect of the target audio device;

[0138] The device adjustment module 206 is configured to query the user feedback instruction of the target audio device if the audio effect does not meet the preset user demand effect, adjust the parameters of the target audio device based on the user feedback instruction, and then return to the step of analyzing the audio effect of the target audio device;

[0139] The target device acquisition module 207 is used to use the target audio device as the audio control device of the voice command if the audio effect meets the preset user requirement effect.

[0140] In detail, the modules described in the audio device mode language control system 200 described in the embodiment of the present invention adopt the same technical means as the audio device mode language control method described in the accompanying drawings when used, and can produce the same technical effects, which will not be repeated here.

[0141] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A method for controlling a mode language of an audio device, characterized in that: The method comprises: Acquire a voice command of a user corresponding to the audio device, extract voice audio of the voice command, and perform energy enhancement on the voice audio to obtain enhanced audio; Analyze the power spectrum of the enhanced audio, perform signal filtering on the enhanced audio based on the power spectrum to obtain filtered audio, calculate the long-time frame power of the filtered audio, and perform noise suppression on the filtered audio based on the long-time frame power to obtain suppressed audio; Querying the multi-channel time-frequency and power of the suppressed audio, performing time-frequency scale adjustment on the channel time-frequency and the power to obtain an adjusted time-frequency and an adjusted power, determining the time-frequency adjusted audio of the suppressed audio based on the adjusted time-frequency and the adjusted power, querying the optimal codebook of the time-frequency adjusted audio, and identifying the audio features of the suppressed audio based on the optimal codebook; Based on the audio feature, parsing the instruction text corresponding to the voice instruction, and based on the instruction text, switching the mode of the audio device to obtain a target audio device; Analyzing the audio effect of the target audio device; If the audio effect does not meet the preset user demand effect, query the user feedback instruction of the target audio device, adjust the parameters of the target audio device based on the user feedback instruction, and then return to the step of analyzing the audio effect of the target audio device; If the audio effect meets the preset user requirement, the target audio device is used as the audio control device of the voice command.

2. The method for controlling the mode language of an audio device according to claim 1, characterized in that: The step of performing energy enhancement on the speech audio to obtain enhanced audio includes: Performing frequency domain conversion on the speech audio to obtain converted audio; Calculating energy levels of different frequency intervals in the converted audio to obtain regional audio energy levels; Determining an energy enhancement interval of the converted audio based on the regional audio energy level; The converted audio is energy enhanced based on the energy enhancement interval to obtain enhanced audio.

3. The method for controlling the mode language of an audio device according to claim 1, characterized in that: The analyzing the power spectrum of the enhanced audio comprises: Extracting an initial signal of the enhanced audio; Calculate the wavelet coefficient spectrum of the initial signal; Based on the wavelet coefficient spectrum, a power spectrum of the enhanced audio is calculated.

4. The method for controlling the audio device mode language according to claim 1, characterized in that: The step of performing signal filtering on the enhanced audio based on the power spectrum to obtain filtered audio includes: Based on the power spectrum, determining a filtering range of the enhanced audio; Based on the filtering range, calculating the signal pass rate of the enhanced audio; Based on the signal pass rate, the signal of the enhanced audio is transformed to obtain filtered audio.

5. The method for controlling the mode language of an audio device according to claim 1, characterized in that: The calculating the long-time frame power of the filtered audio comprises: Performing audio segmentation on the filtered audio to obtain segmented audio; Calculating the signal power of the split audio; Based on the frame power, the long-term frame power of the filtered audio is calculated.

6. The method for controlling the mode language of an audio device according to claim 1, characterized in that: The querying of the optimal codebook for the time-frequency adjusted audio includes: Converting the time-frequency adjusted audio into an audio vector; Performing vector clustering on the audio vector to obtain a clustering vector; Querying the cluster center of the cluster vector; Based on the cluster center, clustering optimization is performed on the cluster vector to obtain an optimized cluster vector; Screening the optimized clustering vector to obtain a screened clustering vector; Based on the screened clustering vectors, an optimal codebook for the time-frequency adjusted audio is determined.

7. The method for controlling the mode language of an audio device according to claim 6, characterized in that: The converting the time-frequency adjusted audio into an audio vector comprises: Performing frame processing on the time-frequency adjusted audio to obtain multiple frames of audio; Performing frequency scale conversion on the multiple frames of audio to obtain converted audio; Perform discrete cosine transform on the converted audio to obtain an audio vector.

8. The method for controlling the mode language of an audio device according to claim 1, characterized in that: The step of identifying the audio features of the suppressed audio based on the optimal codebook includes: Querying characteristic parameters of the audio suppression; Performing vectorization processing on the characteristic parameters to obtain a parameter vector; Matching the parameter vector with the optimal codebook to obtain a matching result; Based on the matching result, the suppressed audio is marked with time-frequency domain features to determine audio features of the suppressed audio.

9. The method for controlling the mode language of an audio device according to claim 1, characterized in that: The analyzing the audio effect of the target audio device includes: Querying control parameters corresponding to the target audio device; Performing an audio effect test on the target audio device according to the control parameters; After the audio effect test is completed, performing intelligent voice interaction with a user corresponding to the target audio device; Based on the interaction result of the intelligent voice interaction, an audio effect of the target audio device is determined.

10. An audio device mode language control system, characterized in that: The system is used to execute the audio device mode language control method according to any one of claims 1 to 9, the system comprising: The audio enhancement module is used to obtain the voice command of the user corresponding to the audio device, extract the voice audio of the voice command, and perform energy enhancement on the voice audio to obtain enhanced audio; an audio suppression module, configured to analyze a power spectrum of the enhanced audio, perform signal filtering on the enhanced audio based on the power spectrum to obtain filtered audio, calculate a long-time frame power of the filtered audio, and perform noise suppression on the filtered audio based on the long-time frame power to obtain suppressed audio; An audio feature recognition module is used to query the multi-channel time-frequency and power of the suppressed audio, perform time-frequency scale adjustment on the channel time-frequency and the power to obtain an adjusted time-frequency and an adjusted power, determine the time-frequency adjusted audio of the suppressed audio based on the adjusted time-frequency and the adjusted power, query an optimal codebook for the time-frequency adjusted audio, and identify the audio features of the suppressed audio based on the optimal codebook; A device mode switching module, used to parse the instruction text corresponding to the voice instruction based on the audio feature, and switch the mode of the audio device based on the instruction text to obtain a target audio device; An audio effect analysis module, used to analyze the audio effect of the target audio device; a device adjustment module, configured to query a user feedback instruction of the target audio device if the audio effect does not meet a preset user requirement effect, adjust parameters of the target audio device based on the user feedback instruction, and then return to the step of analyzing the audio effect of the target audio device; The target device acquisition module is used to use the target audio device as the audio control device of the voice command if the audio effect meets the preset user requirement effect.

Images