Method and system for self-adjusting and identifying smart speakers

Abstract

A method for self-adjusting and identifying smart speakers is provided, which is applied to an audio system. The method includes: pairing smart speakers of the audio system with each other, and collecting the number of signal strength indication values that are not less than a first threshold value to generate a signal strength indication array; generating a feature parameter set based on the signal strength indication array, and a feature parameter including an average value; according to the feature parameter set, using a feature parameter comparison method to identify a first smart speaker; according to the feature parameter set and an identification result of the first smart speaker, using a group identification method to identify other smart speaker. Thereby, this method avoids the user's tedious manual configuration operations and significantly improves the identification efficiency of smart speakers. Comparison of characteristic parameters and signal strength indication values ​​is used to ensure the accuracy and consistency of smart speakers in the identification process and reduce misidentification caused by human interference.

Description

Technical Field

[0001] This invention relates to the field of smart speaker technology, and in particular to a method and system for self-adjustment and identification of smart speaker channels. Prior Technology

[0002] The widespread adoption of wireless smart speakers has provided users with a completely new auditory experience, especially home theater audio systems composed of smart speakers, which further demonstrate the advantages of artificial intelligence in serving family life. However, current methods for configuring wireless speaker channels still rely on manual operation by the user, which presents many inconveniences.

[0003] Taking a 5.1 home theater audio system as an example, the channel information of each smart speaker is typically: center speaker channel, subwoofer channel, left front channel, right front channel, left rear channel, and right rear channel. Correspondingly, each smart speaker with different channel information plays audio signals from different channels. In previous technologies, users needed to manually identify the type of audio system, then match the audio source channels with the speakers one by one, and establish a wireless connection. This configuration method was not only time-consuming and laborious but also placed high demands on the user's operational skills, making it difficult to achieve true intelligence and automation.

[0004] As users increasingly demand greater intelligence and convenience from audio equipment, the familiar manual configuration methods are clearly insufficient to meet their needs. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a method for self-adjustment and identification of smart speaker channels, which aims to solve the problems of complex manual operation, low efficiency and insufficient user experience in the channel configuration methods of smart audio systems in the prior art, and to provide a smart and automated method for self-adjustment and identification of smart speaker channels.

[0006] A first aspect of this invention provides a method for self-adjusting and identifying channels in a smart speaker, applied to an audio system, comprising the following steps:

[0007] Multiple smart speakers in the audio system are paired up, and multiple signal strength indicator values ​​of at least one first threshold value are collected from each pair of smart speakers to generate a signal strength indicator cluster array;

[0008] A set of feature parameters is generated based on multiple signal strength indicator cluster arrays, and the feature parameters include the mean;

[0009] The first smart speaker is identified using a feature parameter comparison method based on the feature parameter set.

[0010] Based on the feature parameter set and the identification results of the first smart speaker, the remaining smart speakers are identified using a group identification method.

[0011] In some embodiments of the first aspect, the first smart speaker is identified using a feature parameter comparison method based on a feature parameter set, including the following steps:

[0012] Sort the feature parameter set according to its mean.

[0013] The two averages of the top two values ​​are selected as the first criterion;

[0014] Based on the comparison results of the first criterion and the feature parameter set, the first smart speaker was identified.

[0015] In some embodiments of the first aspect, based on the feature parameter set and the identification result of the first smart speaker, a group identification method is used to identify the remaining smart speakers, including the following steps:

[0016] The feature parameter set is then removed by removing the feature parameters of the first smart speaker, and the remaining smart speakers are then grouped.

[0017] Within each smart speaker group, the components are sorted according to the magnitude of their characteristic parameters, with every two components of the same value forming a unit.

[0018] Based on the grouping and unit results, the sum of squared differences for each smart speaker is calculated, and the second smart speaker is identified.

[0019] Based on the identification results of the second smart speaker, the remaining smart speakers were identified.

[0020] In some embodiments of the first aspect, based on the feature parameter set and the identification result of the first smart speaker, a group identification method is used to identify the remaining smart speakers, including the following steps:

[0021] From the set of feature parameters, the signal strength indicator value of the first smart speaker is removed, and the two smart speakers with the highest average value between them and the first smart speaker are selected for grouping.

[0022] Within each group of smart speakers, they are sorted according to their mean values, with each pair of means forming a unit in that order.

[0023] Based on the grouping and unit results, the sum of squares of the differences between the two grouped smart speakers is calculated to identify the second smart speaker;

[0024] Based on the identification results of the second smart speaker, the remaining smart speakers were identified.

[0025] In some embodiments of the first aspect, the method further includes comparing the characteristic parameters of the remaining smart speakers with those of the first smart speaker to identify the left smart speaker and the right smart speaker.

[0026] In some embodiments of the first aspect, the method further includes, when the smart speaker device has a microphone array, identifying the left smart speaker and the right smart speaker through the microphone array.

[0027] In some embodiments of the first aspect, the method further includes the following steps:

[0028] Extract the first speaker standard base;

[0029] The characteristic data of each smart speaker in the audio system are normalized to obtain a normalized dataset for each smart speaker;

[0030] Calculate the sum of squared differences between each normalized dataset and the first speaker standard basis;

[0031] When the minimum value among multiple sums of differences equals the sum of differences of the first smart speaker, the channel verification of the first smart speaker is passed.

[0032] In some embodiments of the first aspect, the method further includes the following steps:

[0033] Extract the standard base of the second speaker;

[0034] The absolute values ​​of the feature data of each smart speaker in the audio system after removing the signal strength indication value of the first smart speaker are taken and normalized to obtain the normalized dataset of each smart speaker.

[0035] Calculate the sum of squared differences between each normalized dataset and the second speaker standard basis;

[0036] When the minimum value of the sum of multiple differences equals the sum of differences of the second smart speaker, the channel verification of the second smart speaker is passed.

[0037] In some embodiments of the first aspect, the method further includes the following steps:

[0038] Update the standardized data sets of each smart speaker to the current standard base;

[0039] The audio system calculates the normalized dataset of each smart speaker according to a preset mode, compares it with the current standard, and if the preset conditions are met, the method is re-executed to identify each smart speaker.

[0040] A second aspect of the present invention provides a smart speaker channel self-adjustment identification system for implementing the above-described smart speaker channel self-adjustment identification method.

[0041] The intelligent speaker channel self-adjustment identification method and system of the present invention have the following beneficial effects:

[0042] Simplified operation and improved efficiency: Through intelligent channel self-adjustment and recognition, the user avoids tedious manual configuration operations, significantly improving the efficiency of speaker channel recognition.

[0043] Automation and accuracy: By comparing characteristic parameters and signal strength, the accuracy and consistency of the speaker channel identification process are ensured, reducing erroneous identification caused by human interference.

[0044] Enhance user experience: Achieve efficient identification of multi-channel smart speaker systems, providing users with a smarter and more convenient user experience.

[0045] To further understand the features and technical content of the present invention, please refer to the following detailed description and drawings of the present invention. However, the drawings provided are for reference and illustration only and are not intended to limit the present invention. Simple Explanation of the Diagram

[0046] Figure 1 is a schematic diagram of the channel layout of each smart speaker in a five-channel audio system;

[0047] Figure 2 is a flowchart illustrating a smart speaker channel self-adjustment identification method according to an embodiment of the present invention;

[0048] Figure 3 is a schematic flowchart of an embodiment of the present invention for identifying a first smart speaker using a feature comparison method;

[0049] Figure 4 is a schematic diagram of the process of identifying other smart speakers using the group identification method in the first embodiment of the present invention;

[0050] Figure 5 is a schematic diagram of the process of identifying other smart speakers using the group identification method according to the second embodiment of the present invention;

[0051] Figure 6 is a flowchart illustrating a first smart speaker channel verification method according to an embodiment of the present invention;

[0052] Figure 7 is a flowchart illustrating a second smart speaker channel verification method according to an embodiment of the present invention;

[0053] Figure 8 is a flowchart illustrating a speaker system self-testing method according to an embodiment of the present invention;

[0054] Figure 9 is a schematic diagram of the structure of a smart speaker channel self-adjustment identification system according to an embodiment of the present invention. Implementation

[0055] The following specific embodiments illustrate the implementation of the "Smart Speaker Channel Self-Adjustment Identification Method and System" disclosed in this invention. Those skilled in the art can understand the advantages and effects of this invention from the content disclosed in this specification. This invention can be implemented or applied through other different specific embodiments, and various details in this specification can also be modified and changed based on different viewpoints and applications without departing from the concept of this invention. Furthermore, the accompanying drawings of this invention are for simple illustrative purposes only and are not depictions of actual dimensions; this is stated beforehand. The following embodiments will further describe the relevant technical content of this invention in detail, but the disclosed content is not intended to limit the scope of protection of this invention.

[0056] It should be understood that while terms such as "first," "second," and "third" may be used in this document to describe various components or signals, these components or signals should not be limited by these terms. These terms are primarily used to distinguish one component from another, or one signal from another. Furthermore, the term "or" as used herein should, as appropriate, include any combination of one or more of the associated listed items.

[0057] Figure 1 illustrates the channel layout of the smart speakers in the audio system according to various exemplary embodiments of this application. Taking a typical 5.1 channel audio system as an example, the audio system includes a center channel speaker, two front channel speakers (left front channel speaker and right front channel speaker), two rear channel speakers (left rear left channel speaker and rear right channel speaker), and a subwoofer. The center channel speaker is typically positioned directly in front of the user and is used to primarily present dialogue or main sound. The front left and right channel speakers are positioned on the left and right sides in front of the user, respectively, to present sound effects in the left and right directions. The rear left and right channel speakers are positioned on the left and right sides behind the user to provide surround sound effects. The subwoofer is positioned between the center channel speaker and the front right channel speaker, responsible for playing low-frequency sounds to enhance overall immersion and sound quality. In this embodiment of the invention, a channel refers to an independent audio signal collected or reproduced at different spatial locations during recording or playback. Therefore, the number of channels is the number of audio sources during recording or the corresponding number of speakers during playback. Of course, besides a 5.1-channel audio system, there are other multi-channel audio systems, such as a 7.1-channel audio system, which includes two center channel speakers (left center channel speaker and right center channel speaker) between the two front channel speakers and two rear channel speakers of a 5.1-channel system. In practical implementation, the smart speakers of the audio system pair with the audio source device via wireless connection. The channels of each smart speaker are identified through the self-adjustment identification method of this invention, eliminating the need for manual settings or adjustments by the user, thereby achieving fully automatic channel matching and layout optimization. This layout not only meets the needs of multi-channel sound effects but also simplifies the configuration process of the audio system and improves the user experience. The first distance D1 between the right front channel speaker and the right rear channel speaker is, for example, 4500 mm; the second distance D2 between the right front channel speaker and the left front channel speaker is, for example, 3500 mm; and the third distance D3 between the right rear channel speaker and the user is, for example, 1500 mm.

[0058] Based on the aforementioned application scenarios and audio system architecture, this application provides a method for self-adjusting and identifying smart speaker channels. Applied to an audio system, for multiple independent smart speakers without pre-configured channels, these independent smart speakers can be freely arranged according to the conventional layout of an immersive sound system, without being limited by smart speaker channel information. A signal strength indicator cluster array can be generated by collecting signal strength indicator values ​​between each pair of smart speakers, calculating its characteristic parameters, and comparing them to identify the channel of the first smart speaker. Then, based on the channel identification result of the first smart speaker, a group identification method is used to gradually identify the channel information of the remaining smart speakers. In this way, the multiple independent smart speakers can be combined into an audio system, producing an immersive sound effect when audio signals are played through these multiple smart speakers. This allows for flexible use of these independent smart speakers without speaker channel information, freely combining and configuring the channels of each smart speaker to form an immersive sound system. Therefore, the embodiments of this application can automatically and efficiently identify and match speaker channels, simplify user operation, and improve the intelligence level and user experience of the audio system.

[0059] It should be noted that the executing entity of the smart speaker channel self-adjustment identification method provided in this application embodiment can be a smart speaker (e.g., a smart speaker with control functions or other smart speakers), a functional module and / or functional entity in the smart speaker capable of implementing the method, a terminal device connected to the smart speaker (e.g., a mobile phone, tablet computer, etc.), or a functional module and / or functional entity in the terminal device capable of implementing the method. Furthermore, the execution of this method can also be completed collaboratively by multiple smart speakers in a distributed manner, or through cooperation between the smart speaker and the terminal device. The specific implementation can be determined according to actual usage requirements, and this application embodiment does not impose any limitations on this.

[0060] As shown in Figure 2, this embodiment of the invention provides a smart speaker channel self-adjustment identification method, applied to an audio system, including the following steps:

[0061] S100: Pair multiple smart speakers in the audio system into pairs, and collect multiple signal strength indication values ​​from each pair of smart speakers, which are no less than a first threshold value, to generate a signal strength indication cluster array.

[0062] In this embodiment, smart speakers in an audio system are paired up in pairs using wireless communication technology, and a signal strength indicator cluster array is established based on multiple Received Signal Strength Indicator (RSSI) values ​​between each pair of smart speakers. Each pair of smart speakers pairs up via a wireless communication protocol, for example, Wi-Fi, operating at a frequency such as 5.6 GHz, to collect signal strength. At least a first threshold number of RSSI values ​​are collected between each pair of smart speakers, and these RSSI values ​​generate a signal strength indicator cluster array (hereinafter referred to as a cluster array). Each cluster array is the result of analyzing the relative position and signal quality between the speakers based on the collected RSSI values. This data helps in the subsequent self-adjustment and identification of the smart speaker channels. For example, a 5.1 channel audio system with 6 smart speakers will generate 30 cluster arrays during the transmission and reception pairing process. Each cluster array contains a dataset of multiple collected RSSI values, representing the signal strength information between each pair of smart speakers. It should be noted that, in order to improve stability and accuracy, the number of signal acquisitions for each pair of smart speakers should typically reach a first threshold value. In this invention, the first threshold value is 40 to ensure that the acquired data has sufficient representativeness and statistical significance. This embodiment of the invention uses a first threshold value of 200 acquisitions as an example to collect signal strength indication values.

[0063] It should be noted that the wireless protocol and sampling frequency (i.e., the number of first threshold values) during the signal strength indicator (SSI) acquisition process can be adjusted and optimized according to actual needs. In this scheme, Wi-Fi is used for speaker pairing and signal acquisition. Besides Wi-Fi, Bluetooth, ZigBee, or other wireless protocols can also be used. The SSI is typically represented by the Received Signal Strength Indicator (RSSI). A higher RSSI indicates better signal quality between speakers and a closer distance between devices. In addition, other signal quality parameters (such as SNR, signal-to-noise ratio, and time difference of arrival) can also be used for acquisition.

[0064] In some embodiments, based on the channel reciprocity principle, the number of measurements for the signal strength indicator cluster array of each pair of smart speakers is halved. The channel reciprocity principle states that under ideal symmetrical channel conditions, the signal transmission quality (such as signal strength, noise, etc.) is symmetrical between two communicating devices; that is, the signal transmission quality from device A to device B is the same as the signal transmission quality from device B to device A. In other words, the channel from A to B has the same performance as the channel from B to A under ideal conditions. In this embodiment of the invention, the application of the channel reciprocity principle means that the relative position between speakers can be deduced from the signal strength indicator data, and this data is symmetrical between paired speakers. By collecting the signal strength indicator values ​​between each pair of smart speakers and utilizing the channel reciprocity principle, redundant data can be reduced and computational efficiency improved. For example, in a 5.1 channel audio system, the smart speakers are paired and monitored to obtain a total of 30 signal strength indicator clusters. With the support of the channel reciprocity principle, these clusters can be halved to 15 clusters, further reducing the computational burden and storage pressure of the system, while also ensuring the validity and accuracy of the data.

[0065] S200: Generate a set of feature parameters based on multiple signal strength indicator cluster arrays, including the mean value;

[0066] In this embodiment, statistical analysis of the signal strength indicator cluster array extracts parameters describing the relative positions and signal strength characteristics of each smart speaker, thereby illustrating the identification of smart speakers in an audio system. Specifically, this embodiment processes the signal strength indicator cluster array through data analysis, calculating the average of multiple signal strength indicator values ​​for each cluster array as a feature parameter for each cluster array. The average value of each cluster array represents the average level of signal quality between that pair of smart speakers. The calculation of the average value reflects the stability and reliability of the signal strength between each pair of smart speakers. Thus, even without preset smart speaker channel information, the channel configuration of smart speakers can be identified using the average value as a feature, improving the accuracy and efficiency of self-adjusting channel configuration.

[0067] It should be noted that although the mean is used as the feature parameter in this step of the present invention, it can be extended to other statistical features as needed, such as variance, standard deviation, or, for example, median, skewness, and kurtosis. These features can provide richer information on the signal distribution between smart speakers, further improving the accuracy of identification and enhancing the robustness of data analysis. For example, variance can be used to measure the volatility of signal strength, while standard deviation can reflect the stability of signal strength data. In addition, in practical applications, signal strength acquisition is affected by environmental factors (such as obstacles such as walls and furniture) and signal interference. More complex signal processing methods (such as noise filtering and signal smoothing) can be selected according to the actual situation to further improve signal quality and accuracy.

[0068] S300: Identify the first smart speaker using the feature parameter comparison method based on the feature parameter set;

[0069] In this embodiment, the feature parameters of each smart speaker are compared to classify and identify them according to the feature parameter set. Since the feature parameters can represent information such as the physical location of the smart speaker and the signal characteristics of the sound source, the differences between the feature parameters can be effectively distinguished by comparing the values ​​of the feature parameters or by comparing them with preset threshold values ​​or reference values. This allows the smart speaker that best meets the standard of a specific channel feature (e.g., front, center, rear, etc.) to be identified.

[0070] In some preferred embodiments, S300, based on the feature parameter set, a feature comparison method is used to identify the first smart speaker. The purpose is to sort the smart speakers according to specific statistical characteristics (such as the mean) based on the feature parameter set collected from each smart speaker, so as to further identify the first smart speaker. As shown in Figure 3, an embodiment of the present invention provides a flowchart illustrating the process of identifying the first smart speaker using a feature comparison method, including the following steps:

[0071] S310. Sort the feature parameter set according to the mean value;

[0072] In this embodiment of the invention, the mean represents the average value of multiple signal strength indicators within a cluster array. Each pair of smart speakers in the audio system generates a cluster array, and the multiple signal strength indicators of each cluster array are statistically analyzed to obtain a mean. These means reflect the signal strength level, and thus reflect the distance between a particular smart speaker and other smart speakers.

[0073] Taking a typical 5.1 speaker system as an example, which consists of 6 smart speakers, the above steps can be used to obtain 30 cluster arrays in which the smart speakers are paired with each other. If the reciprocity principle is adopted, the cluster array can be halved to 15 groups (e.g., μ1, μ2, ..., μ15). The characteristic parameters are judged by taking the mean as an example. The mean values ​​of the characteristic parameter set are sorted. These mean values ​​represent the concentration values ​​of the signal strength between the smart speakers.

[0074] The sorting operation in this embodiment can use common sorting algorithms (such as quicksort, bubble sort, etc.) to arrange these averages in order of size. The pair with the largest average in the sorting result represents the pair of smart speakers that receive the most power, which also means that the pair of smart speakers is closest to each other.

[0075] It should be noted that besides using the mean for sorting, other statistical features can also be used to sort cluster arrays. For example: median sorting: For each cluster array, the median of its signal strength can be calculated, and the arrays can be sorted according to the median. The median is more robust than the mean when dealing with extreme values; standard deviation sorting: If signal volatility is a concern, cluster arrays can be sorted based on their standard deviation. Cluster arrays with smaller standard deviations represent more stable signals, while those with larger standard deviations represent more volatile signals; signal-to-noise ratio (SNR) sorting: If there is a lot of interference or noise in the system, the SNR can also be considered as a sorting criterion. These alternatives can be selected according to specific application scenarios and requirements to further improve the system's identification accuracy and flexibility.

[0076] S320. Select the two mean values ​​(maximum and second largest values) of the top two values ​​from the feature parameter set as the first criterion;

[0077] The aim is to identify and determine smart speakers by further filtering the set of characteristic parameters in an audio system. In a multi-channel audio system, by sorting all the means and selecting the top two means, the smart speaker most likely to meet the matching condition can be identified based on the difference in the means and designated as the first smart speaker.

[0078] Select the two mean values ​​from all sorted mean values, namely mean 15[0] and mean 15[1], as the first criterion. The selection of the characteristic parameters of the first criterion represents the two sets of smart speakers with the strongest signal strength between them. Usually, the relative position of the smart speakers that meet the first criterion can determine the specific position conditions.

[0079] S330. Based on the comparison results of the first criterion and the feature parameter set, the first smart speaker is identified.

[0080] Since the mean represents the main characteristic of the signal strength indication value between smart speakers, and the cluster array is generated by collecting the signal strength indication value between multiple pairs of smart speakers, each element in the cluster array represents the signal strength indication value between a smart speaker and another smart speaker. By comparing the mean of these signal indication values, it is possible to determine which smart speakers' positions match the first criterion, thereby accurately identifying the first smart speaker.

[0081] In the 5.1 audio system shown in Figure 1, the subwoofer usually has a unique signal characteristic, and its signal strength indication value satisfies the sorting of the first two average values. Therefore, the subwoofer is determined to be the first intelligent speaker, that is, the subwoofer in the 5.1 audio system.

[0082] It should be noted that the method of this embodiment is not only applicable to 5.1 audio systems, but can also be effectively applied to more complex audio systems, such as 7.1 audio systems and 9.1 audio systems. The subwoofers in these systems meet the characteristic parameters that rank first and second in terms of average value as the first criterion. Thus, the first smart speaker, i.e. the subwoofer, can be determined through the method of this embodiment.

[0083] S400: Based on the feature parameter set and the identification results of the first smart speaker, the group identification method is used to identify the remaining smart speakers.

[0084] In this embodiment of the invention, the group identification method is a technique for grouping the remaining smart speakers in an audio system according to their characteristic parameters. Each group represents a class of smart speakers with similar characteristic parameters. By grouping the cluster array, the characteristics of each group of smart speakers are further analyzed. The grouping is based on the differences between the signal strength indicators, relative positions, and other characteristic parameters of the smart speakers. For example, in a 5.1 audio system, the front speakers, surround speakers, subwoofers, and speakers located on the left or right sides will exhibit different characteristic parameters. Through grouping, they can be classified into different categories. Based on the grouping, and referring to the identification results of the first smart speaker, the channel roles of the remaining smart speakers are further identified. Specifically, the group identification method uses the characteristic parameters between the characteristic parameter sets to divide the smart speakers into different groups, and combines the characteristic parameter relationships between each group and the first smart speaker to identify the channels of the other smart speakers. The identification result of the first smart speaker refers to the identification result of the first smart speaker, such as the subwoofer, through comparison and analysis in the above steps. The identification result of the first smart speaker provides a basis for the subsequent identification of other smart speakers.

[0085] In some preferred embodiments, S400, based on the feature parameter set and the identification result of the first smart speaker, the remaining smart speakers are identified using a group identification method. As shown in Figure 4, the first embodiment of the present invention provides a flowchart of identifying the remaining smart speakers using a group identification method, including the following steps:

[0086] S411. From the feature parameter set, remove the feature parameters corresponding to the first smart speaker, and then group the remaining unidentified smart speakers;

[0087] In this embodiment, the mean value corresponding to the first smart speaker is removed from all mean values ​​of the audio system to avoid interference from the first smart speaker in the subsequent smart speaker grouping process. The remaining smart speakers will be classified and grouped based on their signal characteristics or after statistical analysis of the cluster array.

[0088] Taking a 5.1 speaker system as an example, there are a total of 6 smart speakers, including two front speakers, two surround speakers, one subwoofer, and one center speaker. After removing the characteristic parameters corresponding to the first smart speaker (such as the subwoofer identified by the above method), the characteristic parameters corresponding to the remaining 5 smart speakers are grouped on a per-speaker basis. This results in a grouping of the 5 smart speakers. This grouping method can better determine the channel of each smart speaker by the relative differences in the characteristic parameters of the smart speakers.

[0089] S412. The feature parameters within each group of smart speakers are sorted according to their mean values, and every two mean values ​​are grouped into one unit according to the sorting order;

[0090] In this embodiment of the invention, multiple feature parameters within each group of smart speakers are further divided into units. In actual implementation, each smart speaker can first be sorted according to a certain feature parameter, such as the signal strength indicator value. The signal strength indicator value (e.g., RSSI value) is usually used to represent the strength of a wireless signal; the stronger the signal, the larger the value. Therefore, the purpose of the sorting process is to separate the stronger signals from the weaker signals in the currently grouped smart speakers, providing a clear distinction for subsequent identification.

[0091] Taking a 5.1 speaker system as an example, the sorted means are divided into two units for more detailed classification and identification. First, the four means (mean[0], mean[1], mean[2], mean[3]) corresponding to each smart speaker are sorted according to the size sorting rules. After sorting, each pair of means is used as a unit. For example, the two largest and second largest means (mean[0], mean[1]) after sorting are grouped into one unit, and the two smallest and second smallest means (mean[2], mean[3]) after sorting are grouped into another unit. This helps to further refine the signal data of each smart speaker and provides more stable data support for subsequent grouping and matching identification.

[0092] It should be noted that the grouping and sorting methods and criteria can be adjusted according to different application scenarios and the requirements of the audio system. For example, the sorting rules can be automatically adjusted based on the specifications of different audio systems or the actual signal strength in different environments. In other application scenarios, in addition to signal strength values, other characteristic parameters (such as signal stability, transmission delay, etc.) can also be used for sorting and grouping. Therefore, this invention does not impose specific restrictions on the sorting rules and grouping strategies, and they can be adjusted according to actual needs.

[0093] S413. Based on the grouping results and unit results, calculate the sum of squares of differences for each smart speaker, and identify the second smart speaker;

[0094] This invention employs a sum-of-squares (SFS) analysis method (i.e., calculating the sum of squares of the differences between two numbers) to quantify the differences between signal strengths or characteristic parameters, and to identify the channel role of a smart speaker based on these differences. Through this method, the invention can accurately identify the speaker group with the smallest difference from other speakers within a given group, and can determine the channel information of a given group.

[0095] The sum of squares (J) is a method for calculating the differences within groups of different smart speakers. In this step, the formula for calculating the sum of squares (J) is:

[0096] J=

[0097] Where mean[0], mean[1], mean[2], and mean[3] are the four means within each group of smart speakers. For example, mean[0] and mean[1] represent a group with a larger mean, and mean[2] and mean[3] represent a group with a smaller mean. This formula quantifies the differences between different smart speakers by calculating the sum of the squares of the differences between each pair of means. In this way, a sum of squares (J value) can be calculated for each smart speaker, and these J values ​​represent the relative degree of difference of the smart speaker. The J values ​​of each smart speaker (such as J1, J2, J3, J4, J5) are obtained. Finally, the smart speaker with the smallest J value is selected as the second smart speaker. In this embodiment of the invention, the second smart speaker is the center speaker.

[0098] In an audio system, as shown in Figure 1, the center speaker is typically positioned in the center of the front speakers, and in a 5.1 or 7.1 audio system, it usually exhibits a significant difference in signal position compared to other smart speakers (such as front and rear speakers). Due to its unique position, the center speaker experiences less signal variation within its speaker unit compared to other speakers (such as front or surround speakers), resulting in the lowest J-value, making it suitable as the center speaker for this smart system.

[0099] It should be noted that the method for calculating the J value in this embodiment of the invention is only one implementation scheme. In different application scenarios, the calculation method of the sum of differences can be adjusted according to actual needs. For example, other statistics (such as variance, standard deviation, etc.) can be chosen to replace the sum of differences, or weighting factors can be introduced on the basis of the sum of differences to assign different weights to the signal characteristics of different speakers.

[0100] S414. Based on the identification result of the second smart speaker, further identify the remaining smart speakers.

[0101] In this embodiment of the invention, by further analyzing the signal characteristics of the second smart speaker (such as the center speaker channel), the channels of the other smart speakers can be determined based on the differences in characteristic parameters between the second smart speaker and other unidentified smart speakers. This step combines the sorting and differences of the smart speakers to further classify the position and channel information of the other smart speakers, thereby accurately identifying the smart speakers.

[0102] After identifying the second smart speaker, i.e., the center speaker channel, the smart speakers are sorted according to the magnitude of their characteristic parameters relative to the other unidentified smart speakers. The two smart speakers corresponding to the two largest and second largest characteristic parameters (e.g., the mean) represent the characteristics of the front speakers, while the other two smart speakers corresponding to the two smallest and second smallest characteristic parameters (e.g., the mean) represent the characteristics of the rear speakers. In a 7.1 speaker system, the two smart speakers corresponding to the two middle characteristic parameters (e.g., the mean) represent the characteristics of the center speaker.

[0103] In some preferred embodiments, S400, based on the feature parameter set and the identification result of the first smart speaker, the remaining smart speakers are identified using a group identification method. As shown in Figure 5, a flowchart illustrating the process of identifying remaining smart speakers using a group identification method according to a second embodiment of the present invention is provided, including the following steps:

[0104] S421. From the characteristic parameter values, remove the signal strength indication value of the first smart speaker, and select the two smart speakers with the highest average value among them and the first smart speaker for grouping;

[0105] In this embodiment of the invention, similar to step S411, the feature parameters corresponding to the first smart speaker are removed, and the two smart speakers with the highest average value between them and the first smart speaker are selected, as shown in Figure 1. In an audio system, the two smart speakers closest to the subwoofer must include the center speaker; therefore, only the two smart speakers closest to the subwoofer need to be selected for subsequent identification. This method can significantly reduce the number of cluster arrays that need to be processed, thereby improving the efficiency of the identification process. This method is not only applicable to 5.1 or 7.1 systems, but can also be extended to audio systems with more channels, such as 9.1 or 11.1 systems. In different audio systems, selecting the two smart speakers closest to the first smart speaker for identification can effectively reduce computational complexity.

[0106] S422. Within each group of smart speakers, sort them according to their mean values, and group them into units with two mean values ​​in the sorting order.

[0107] In this embodiment of the invention, similar to S421, it is only necessary to sort the mean values ​​of the two smart speakers that have already been grouped, and then further divide the sorted results into two units.

[0108] S423. Based on the grouping results and unit results, calculate the sum of squares of the differences between the two smart speakers to identify the second smart speaker;

[0109] In this embodiment of the invention, similar to S421, the second smart speaker, i.e. the center speaker, can be identified simply by calculating the sum of squared differences between the two smart speakers.

[0110] S424. Based on the identification result of the second smart speaker, the remaining smart speakers are identified.

[0111] In this embodiment of the invention, similar to S424, the front speaker, rear speaker and other speakers (such as the center speaker) can be accurately identified based on the identification result of the second smart speaker.

[0112] In some embodiments, the smart speaker channel self-adjustment identification method further includes comparing the characteristic parameters of the other smart speakers with those of the first smart speaker to identify the left smart speaker and the right smart speaker.

[0113] In this embodiment of the invention, the channel type (i.e., left channel and right channel) of the smart speaker is determined by comparing the characteristic parameters of the other smart speakers with those of the first smart speaker. In specific implementation, the characteristic parameters of each smart speaker are compared with those of the first smart speaker* (e.g., a subwoofer). The one with the smaller mean value is the left smart speaker, and the one with the larger mean value is the right smart speaker. By comparing the mean values ​​of each smart speaker with those of the first smart speaker, two front smart speakers (left front smart speaker and right front smart speaker) and two rear smart speakers (left rear smart speaker and rear smart speaker) can be identified. In a 7.1 audio system, two center smart speakers (left center smart speaker and right center smart speaker) can also be identified.

[0114] In some embodiments, when the smart speaker device has a microphone array, the left and right smart speakers are identified through the microphone array. In this embodiment of the invention, a microphone array refers to an array composed of a plurality of microphone elements, through which sound source signals are received at different positions and angles. Each microphone independently collects sound data, and the system can determine spatial information such as the direction and distance of the sound source based on the time difference of arrival (TDOA) between the plurality of microphones. Using an array of multiple microphones can more accurately locate the direction of the sound source than a single microphone, and can identify and process complex sound field information. For example, when sound comes from the left, the microphone closer to the left in the microphone array will receive the sound first, while the microphone closer to the right will receive the sound later. By calculating the time difference, the angle of the sound source can be determined, thereby identifying the left and right smart speakers.

[0115] It should be noted that the microphone array in the embodiments of the present invention is not limited to a specific number or type of microphones; any microphone arrangement that can provide time or spatial differences can be applied to the present invention. For example, the number of microphones, their arrangement, and the signal processing methods can be flexibly adjusted according to specific applications. The method for identifying the left and right channels is not limited to the time difference of the microphone array; it can also be combined with other acoustic characteristics, such as signal amplitude and frequency response, for comprehensive analysis. In addition, for some devices that do not support complex microphone arrays, directional microphones can be used. These microphones can receive sound signals from a specific direction, and by analyzing the direction of the received audio signals, the speaker channels can be further determined.

[0116] In some embodiments, the smart speaker channel self-adjustment identification method further includes a smart speaker channel verification method. This method verifies the channel type by calculating the sum of squared differences between the characteristic data of each smart speaker channel and a standard base, ensuring accurate channel identification for each smart speaker in the audio system. In this invention, the standard base is a preset set of reference data representing the characteristic parameters of smart speaker channels (e.g., subwoofer channels, center speaker channels, etc.). The standard base is typically obtained by collecting and analyzing various performance parameters (such as signal strength, frequency response, and latency) of the smart speaker under normal operating conditions. The verification method uses normalization to eliminate physical differences between different speakers, making the comparison of characteristic data more accurate and consistent. By determining the minimum value of the sum of squared differences, the speaker closest to the standard base can be effectively identified, and channel verification can be performed using that speaker.

[0117] In some embodiments, as shown in FIG6, this embodiment of the invention provides a schematic diagram of the implementation process of a channel verification method for a first smart speaker. The smart speaker channel self-adjustment identification method further includes a channel verification method for the first smart speaker, comprising the following steps:

[0118] S510, Extract the first speaker standard base;

[0119] In this embodiment of the invention, the standard basis of the first speaker is a preset set of characteristic parameters of a subwoofer and other speakers. If the characteristic parameters use the mean, then the standard set of the first speaker is:

[0120] ={μ1,μ2,…μ5}

[0121] Where μ represents the average signal strength indication value measured between the first speaker and other speakers in pairs.

[0122] S520. Normalize the feature data of each smart speaker in the audio system to obtain a normalized dataset for each smart speaker;

[0123] In this embodiment of the invention, the normalization process preprocesses the feature data of each smart speaker to convert the signal characteristics of different smart speakers in the audio system into the same scale or range. Since the signal characteristics of each smart speaker may differ, the normalization process can unify these signal characteristics to the same standard scale, thereby eliminating differences between devices and facilitating comparison and analysis.

[0124] The specific method of normalization can be to subtract the mean of the feature data of each smart speaker and divide it by its standard deviation, or to directly normalize through the maximum and minimum values, etc. The embodiments of the present invention do not make specific limitations.

[0125] The result of normalizing the feature data of a certain smart speaker is

[0126] ={μ1,μ2,…μ5}

[0127] Similarly, the result of normalizing the feature data of other smart speakers can be obtained. , , and .

[0128] S530. Calculate the sum of squared differences between each normalized dataset and the first speaker standard basis;

[0129] In this embodiment of the invention, the sum of differences is used to measure the difference between two datasets (or features). Specifically, the sum of differences between the normalized dataset of each smart speaker and the first speaker standard basis is calculated using the following formula:

[0130] J1=

[0131] in, For each item in a standardized dataset of a specific smart speaker, 4 represents each item in the first speaker standard set. In this embodiment, the first speaker standard set is the subwoofer standard set, and 4 represents the number of data items.

[0132] Similarly, the differences and sums of other smart speakers, J2, J3, J4, J5, and J6, can be calculated.

[0133] S540. When the minimum value among multiple sums of differences is the sum of differences of the first smart speaker, then the channel verification of the first smart speaker is passed.

[0134] In this embodiment of the invention, by comparing the sum of squared differences between all smart speakers and the standard basis of the first speaker, the smart speaker with the smallest difference is selected, and its feature data is considered to be closest to that of the first smart speaker, thus determining that the smart speaker is the first smart speaker. Specifically, if the J value of the subwoofer obtained through the above method in the audio system is the smallest, then the first smart speaker determined in the smart speaker channel self-adjustment identification method is confirmed, and the channel verification of the first smart speaker is performed.

[0135] In some embodiments, as shown in FIG7, this embodiment of the invention provides a schematic diagram of the implementation process of a channel verification method for a second smart speaker. The smart speaker channel self-adjustment identification method further includes a channel verification method for a second smart speaker, comprising the following steps:

[0136] S610, Extract the standard base of the second speaker;

[0137] In this embodiment of the invention, the standard base of the second speaker is a preset set of feature parameters of the center speaker and other smart speakers. If the feature parameters use the mean, then the standard set of the second speaker is:

[0138] ={μ1,μ2,…μ5}

[0139] Where μ represents the average signal strength indication value measured when the second speaker is paired with other speakers.

[0140] S620. After removing the signal strength indication value of the first smart speaker, the absolute values ​​of the feature data of each smart speaker in the audio system are taken and normalized to obtain the normalized dataset of each smart speaker;

[0141] In this embodiment of the invention, similar to S520, the feature data of each smart speaker in the audio system after removing the signal strength indication value of the first smart speaker is preprocessed by normalization.

[0142] S630. Calculate the sum of squared differences between each normalized dataset and the second speaker standard basis;

[0143] In this embodiment of the invention, the sum of squared differences (SSD) is used to measure the difference between two datasets (or features). Specifically, the SSD between the normalized dataset of each smart speaker and the second speaker standard basis is calculated using the following formula: J 1 .

[0144] in, For each item in a standardized dataset of a specific smart speaker, For each item in the second speaker standard base, in this embodiment the first speaker standard set is the subwoofer standard base, and 3 is the number of data items after removing the first smart speaker.

[0145] Similarly, the sum of differences J2, J3, J4, and J5 of other smart speakers can be calculated.

[0146] S640. When the minimum value among multiple sums of differences is the sum of differences of the second smart speaker, then the channel verification of the second smart speaker is passed.

[0147] In this embodiment of the invention, by comparing the sum of squared differences between all smart speakers and the standard basis of the second speaker, the speaker with the smallest difference is selected, and its feature data is considered to be closest to that of the second smart speaker, thus determining that the speaker is the second smart speaker. Specifically, if the J value of the center speaker obtained through the above method in the audio system is the smallest, then the second smart speaker identified in the smart speaker self-adjustment identification method is confirmed, and the channel verification of the second smart speaker is passed.

[0148] In some embodiments, the smart speaker channel self-adjustment identification method further includes a speaker system self-test method. The self-test function of the audio system is primarily used to automatically adapt and update the system configuration when the smart speaker's location is changed or altered. By updating the normalized dataset to the current standard base, the system can determine the current standard base and thus detect changes in device status based on it, responding promptly and re-identifying the smart speaker channels. The self-test method ensures that the system can still efficiently and accurately identify channel information during changes.

[0149] As shown in Figure 8, this embodiment of the invention provides a schematic diagram of the implementation process of a speaker system self-test method. The speaker system self-test method includes the following steps:

[0150] S710, Update the standardized datasets of each smart speaker to the current standard base;

[0151] In this embodiment of the invention, the normalized dataset of each smart speaker is compared and updated with the standard base of the current audio system. The standard base typically represents the normal operating state and characteristic data of the audio system, including the aforementioned first speaker standard base and second speaker standard base, or the standard base of all smart speaker channels. The standard base data may include characteristics such as the speaker's channel layout, signal strength indication, and frequency response.

[0152] S720: The audio system calculates the normalized dataset of each smart speaker according to a preset mode, compares it with the current standard, and if the preset conditions are met, the method described above is re-executed to identify each smart speaker.

[0153] In this embodiment of the invention, the preset mode refers to the audio system updating the speaker feature data according to certain rules or cycles during self-adjustment and identification. The preset mode can include the following methods: calculating the normalized dataset of each smart speaker upon startup; recalculating speaker data and performing identification each time the audio system is powered on; periodically checking the normalized dataset of each smart speaker; the system performs identification and update operations according to a set time interval (e.g., daily, weekly); and frequency-based normalized dataset of each smart speaker; updating data and identifying channels periodically based on the usage frequency of the audio system. Through the preset mode, it can be ensured that the audio system always maintains the latest smart speaker channel configuration. The preset condition refers to a difference threshold value set when comparing the normalized dataset of each smart speaker channel in the audio system with the current standard base. If the difference between the normalized dataset and the current standard base exceeds a certain preset threshold value, the channel identification method needs to be re-executed. Specifically, if the difference exceeds the second threshold (e.g., 0.1), it indicates that the configuration or characteristics of the audio system have changed significantly, and the smart speaker channel self-adjustment identification method needs to be re-executed for re-identification.

[0154] As shown in Figure 9, this embodiment of the invention provides a smart speaker channel self-adjustment identification system for implementing the above-mentioned smart speaker channel self-adjustment identification method. The system includes a signal acquisition module M100 and a control module M200.

[0155] The signal acquisition module M100 is configured to acquire the signal strength indication values ​​of each smart speaker in the audio system and transmit the data to the control module;

[0156] The control module M200 is configured to schedule the operation of various functional modules and units, and to perform processing tasks based on the acquired signal strength indication data. The control module can contain multiple functional units to control the execution of various steps in the speaker channel identification method.

[0157] Preferably, the control module M200 includes different types of databases, such as relational databases and NoSQL databases, to expand its applicability and functionality. Relational databases can be used to store structured data, while NoSQL databases can be used to store unstructured data, supporting more diverse data storage needs. This design provides the smart speaker channel self-adjustment recognition system with more flexible and comprehensive data management and query functions, thereby improving system performance and efficiency.

[0158] In the smart speaker channel self-adjustment identification system of the present invention, the functions of each module can be implemented using the specific implementation of the smart customer service method described above, which will not be elaborated here.

[0159] This invention also provides a smart speaker channel self-adjustment identification program product, including computer-executable instructions, which, when executed by a processor, implements the aforementioned smart speaker channel self-adjustment identification method.

[0160] [Beneficial Effects of the Examples]

[0161] One of the beneficial effects of this invention is that the smart speaker channel self-adjustment identification method and system provided by this invention simplifies operation and improves efficiency, avoiding tedious manual configuration operations for users and significantly improving the efficiency of speaker channel identification. It ensures the accuracy and consistency of the speaker channel identification process, reducing erroneous identification caused by human interference. It achieves efficient identification of multi-channel smart audio systems, providing users with a smarter and more convenient user experience.

[0162] The content disclosed above is only a preferred and feasible embodiment of the present invention, and is not intended to limit the scope of the patent application of the present invention. Therefore, all equivalent technical changes made using the contents of the present invention specification and drawings are included in the scope of the patent application of the present invention.

[0163] D1: First Distance D2: Second distance D3: Third Distance S100-S400, S310-S330, S411-S414, S421-S424, S510-S540, S610-S640, S710-S720: Steps M100: Signal Acquisition Module M200: Control Module

Claims

1. A method for identifying the self-adjustment of a smart speaker channel, applied to an audio system, the method comprising the following steps: pairing multiple smart speakers in the audio system into pairs, and collecting multiple signal strength indication values ​​of each pair of smart speakers (not less than a first threshold value) to generate a signal strength indication cluster array; generating a feature parameter set based on the multiple signal strength indication cluster arrays, the feature parameter set containing multiple feature parameters, each feature parameter including a mean value; identifying a first smart speaker using a feature parameter comparison method based on the feature parameter set; and identifying the remaining smart speakers using a grouping identification method based on the feature parameter set and the identification result of the first smart speaker.

2. The smart speaker channel self-adjustment identification method as described in claim 1, wherein the first smart speaker is identified by the feature parameter comparison method based on the feature parameter set, includes the following steps: sorting the feature parameter set according to the magnitude of multiple mean values; The average of the two values ​​ranked first and second is selected as a first criterion; The first smart speaker is identified based on the comparison result between the first criterion and the feature parameter set.

3. The smart speaker channel self-adjustment identification method as described in claim 2, wherein the identification of other smart speakers using the group identification method based on the feature parameter set and the identification result of the first smart speaker includes the following steps: removing multiple feature parameters corresponding to the first smart speaker from the feature parameter set, and then grouping the remaining unidentified smart speakers; sorting the smart speakers within each group according to their mean values, with each group consisting of two speakers in order of sorting; calculating the sum of squares of differences for each smart speaker based on a grouping result and a unit result to identify a second smart speaker; and identifying the remaining smart speakers based on the identification result of the second smart speaker.

4. The smart speaker channel self-adjustment identification method as described in claim 2, wherein the identification of the remaining smart speakers using a group identification method based on the feature parameter set and the identification result of the first smart speaker includes the following steps: removing multiple feature parameters corresponding to the first smart speaker from the feature parameter set, and selecting the two smart speakers with the highest average value between them and the first smart speaker for grouping; sorting the smart speakers within each group according to their average value, with each pair forming a unit according to the sorting order; calculating the sum of squares of the differences between the two smart speakers in each group based on the grouping result and the unit result, and identifying the second smart speaker; and identifying the remaining smart speakers based on the identification result of the second smart speaker.

5. The smart speaker channel self-adjustment identification method as described in claim 3 or 4 further includes: comparing the average value between each of the smart speakers and the first smart speaker to identify a left smart speaker and a right smart speaker.

6. The smart speaker channel self-adjustment identification method as described in claim 3 or 4 further includes: when the smart speaker has a microphone array, identifying a left smart speaker and a right smart speaker through the microphone array.

7. The smart speaker channel self-adjustment identification method as described in claim 1 further includes: extracting a first speaker standard basis; performing normalization processing on the feature data of each smart speaker in the audio system to obtain a normalized data set of each smart speaker; calculating the sum of squared differences between each normalized data set and the first speaker standard basis; and when the minimum value among the plurality of sums of squared differences is the sum of squared differences of the first smart speaker, then the channel verification of the first smart speaker is passed.

8. The smart speaker channel self-adjustment identification method as described in claim 1 further includes: extracting a second speaker standard basis; normalizing the absolute value of the feature data of each smart speaker in the audio system after removing multiple signal strength indication values ​​corresponding to the first smart speaker to obtain a normalized data set for each smart speaker; calculating the sum of squared differences between each normalized data set and the second speaker standard basis; and passing the channel verification of the second smart speaker if the minimum value among the multiple sums of squared differences is the sum of squared differences of the second smart speaker.

9. The smart speaker channel self-adjustment identification method as described in claim 7 or 8 further includes: updating the normalized data set of each smart speaker to a current standard base; the audio system calculates the normalized data set of each smart speaker according to a preset mode, compares it with the current standard base, and if a preset condition is met, then re-executes the smart speaker channel self-adjustment identification method to identify each smart speaker.

10. A smart speaker channel self-adjustment identification system for performing the smart speaker channel self-adjustment identification method described in any one of requests 1 to 9.

Images