A method and system for ambient adaptive volume adjustment for a screen knob device

Abstract

The application provides an environment adaptive volume adjustment method and system of a screen knob device, relates to the technical field of data processing, and obtains an environment noise signal, a knob operation amount, a current output volume, a display screen wake-up state and a near-field operation state; divides the environment noise signal into a sampling window, generates a sampling window feature set, removes a transient impact window to generate an effective noise window set; determines a target continuous noise window group according to a low-frequency energy proportion and sampling time continuity, and generates a continuous window number and a continuous noise net value; generates an automatic output volume when a volume adjustment condition is met; generates a near-field misamplification event and a volume reduction amount according to a volume reduction operation and a near-field operation state within a preset correction duration after the automatic output volume is formed; generates a misadjustment scene fingerprint according to the corresponding effective noise window set and the volume reduction amount; and when a subsequent effective noise window set and the misadjustment scene fingerprint are matched, corrects an automatic volume increment and adjusts the current output volume.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to an environmentally adaptive volume adjustment method and system for a rotary device with a screen. Background Technology

[0002] Existing technology typically integrates a rotary encoder, push-button switch, display driver module, and audio control module into a knob device with a display screen. When the user rotates the knob, the controller generates volume increase / decrease commands based on the direction and number of pulses, and simultaneously displays the volume level, percentage, or icon on the screen. Some devices also collect ambient sound pressure levels via a microphone, automatically increasing the playback volume when the ambient noise exceeds a preset threshold and restoring the default volume when it falls below the threshold.

[0003] When using a rotary knob with a screen to control a smart speaker or central control device in the kitchen, the range hood generates continuous low-frequency noise when it starts, while chopping vegetables and the sound of a spatula hitting the stove produce short-term spike noise. If current technology adjusts the volume based solely on instantaneous sound pressure level and a fixed threshold, the device may misinterpret the moment the spatula hits the stove as a prolonged increase in ambient noise and suddenly amplify the voice prompt. After the range hood is turned off, the lack of judgment on the type and duration of noise may cause the volume to not drop in time, resulting in users hearing excessively high volumes when operating the knob at close range. Summary of the Invention

[0004] The purpose of this invention is to provide an environmentally adaptive volume adjustment method and system for a device with a screen and knob, aiming to solve the problems mentioned in the background art.

[0005] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows: A first aspect includes an environmentally adaptive volume adjustment method for a device with a screen and knob, the method comprising: Acquire the ambient noise signal, knob operation amount, current output volume, display wake-up status and near-field operation status of the device with screen and knob, wherein the near-field operation status is used to characterize whether the user has entered the operation distance of the device with screen and knob. The environmental noise signal is continuously divided into sampling windows. The average sound pressure level, peak sound pressure level, peak duration and low frequency energy ratio of each sampling window are calculated to generate a sampling window feature set. Based on the feature set of the sampling windows, the sampling windows are sorted by time. Sampling windows whose difference between peak sound pressure value and average sound pressure value exceeds a preset impact difference threshold and whose peak duration is less than a preset impact duration threshold are marked as transient impact windows. Transient impact windows are then removed to generate an effective noise window set. Low-frequency filtering is performed based on the proportion of low-frequency energy in the effective noise window set, and the groups are formed by sampling time continuity to obtain the target continuous noise window group; the number of continuous windows and the net continuous noise value are generated based on the number of sampling windows in the target continuous noise window group and the average sound pressure value of each sampling window. When the net continuous noise value exceeds the preset environmental noise threshold and the number of continuous windows reaches the preset continuous window number threshold, the automatic volume increment is determined based on the difference between the net continuous noise value and the preset environmental noise threshold, and the automatic output volume is generated based on the sum of the current output volume and the automatic volume increment, wherein the automatic output volume does not exceed the maximum output volume allowed by the device. Adjust the current output volume to the automatic output volume, and within the preset correction time after the automatic output volume is formed, when the knob operation indicates a decrease in volume and the near-field operation state is within the operation distance of the knob device with screen, generate a near-field false amplification event and a volume reduction amount. Generate a mis-adjustment scene fingerprint based on the set of effective noise windows corresponding to near-field mis-amplification events and the amount of volume reduction. When the subsequent set of effective noise windows matches the fingerprint of the misadjusted scene, the corresponding automatic volume increment is corrected according to the volume reduction amount, and the current output volume is adjusted according to the corrected automatic volume increment.

[0006] Secondly, an environmentally adaptive volume control system for a rotary knob device with a screen, the system comprising: The data acquisition module is used to acquire the environmental noise signal, knob operation amount, current output volume, display wake-up status and near-field operation status of the screen-equipped rotary device, wherein the near-field operation status is used to characterize whether the user has entered the operation distance of the screen-equipped rotary device. The sampling window feature generation module is used to continuously divide the environmental noise signal into sampling windows, calculate the average sound pressure value, peak sound pressure value, peak duration and low frequency energy ratio of each sampling window, and generate a sampling window feature set. The transient impact rejection module is used to sort the sampling windows by time according to the sampling window feature set, mark the sampling windows whose peak sound pressure value and average sound pressure value difference exceeds the preset impact difference threshold and whose peak duration is less than the preset impact duration threshold as transient impact windows, and reject the transient impact windows to generate an effective noise window set. The continuous noise generation module is used to perform low-frequency filtering based on the proportion of low-frequency energy in the effective noise window set, and to group them according to the continuity of sampling time to obtain the target continuous noise window group; based on the number of sampling windows in the target continuous noise window group and the average sound pressure value of each sampling window, the number of continuous windows and the net continuous noise value are generated. An automatic volume generation module is used to determine an automatic volume increment based on the difference between the net continuous noise value and the preset environmental noise threshold when the net continuous noise value exceeds a preset environmental noise threshold and the number of continuous windows reaches a preset continuous window number threshold, and to generate an automatic output volume based on the sum of the current output volume and the automatic volume increment, wherein the automatic output volume does not exceed the maximum output volume allowed by the device. The near-field false amplification event generation module is used to adjust the current output volume to the automatic output volume, and within the preset correction time after the automatic output volume is formed, when the knob operation indicates a decrease in volume and the near-field operation state is within the operating distance of the knob device with screen, a near-field false amplification event and a volume reduction amount are generated. The mis-amplification scene fingerprint generation module is used to generate a mis-amplification scene fingerprint based on the effective noise window set and volume reduction amount corresponding to the near-field mis-amplification event. The volume adjustment module is used to correct the corresponding automatic volume increment based on the volume reduction amount when the subsequent effective noise window set matches the fingerprint of the misadjusted scene, and adjust the current output volume based on the corrected automatic volume increment.

[0007] The above-described solution of the present invention has at least the following beneficial effects: This invention acquires ambient noise signals, knob operation amounts, current output volume, display screen wake-up status, and near-field operation status, enabling the volume adjustment process to simultaneously consider both ambient noise and user near-field operation. The near-field operation status indicates whether the user has entered the operating distance of the knob-equipped device with a screen, allowing subsequent volume reduction operations to establish a correspondence with the user's near-field operation scenario.

[0008] Next, this invention continuously divides the environmental noise signal into sampling windows and calculates the average sound pressure level, peak sound pressure level, peak duration, and low-frequency energy ratio, ensuring that each sampling window possesses sound pressure intensity, impact characteristics, and low-frequency characteristics. Transient impact windows are identified and eliminated by using the difference between the peak and average sound pressure levels and the peak duration, preventing short-duration spikes such as those from chopping vegetables or colliding with a spatula from participating in subsequent continuous noise assessment, thus reducing the likelihood of sudden amplification of voice prompts due to short-duration impact noise.

[0009] Furthermore, this invention performs low-frequency screening based on the proportion of low-frequency energy in the effective noise window set, and groups them according to the continuity of sampling time to obtain a target continuous noise window group; then, based on the number of sampling windows in the target continuous noise window group and the average sound pressure value of each sampling window, it generates the number of continuous windows and the net continuous noise value. Thus, continuous low-frequency noise from range hoods and other sources can be distinguished from short-term spike noise, allowing automatic volume adjustment to be constrained by noise type, noise intensity, and duration simultaneously.

[0010] Furthermore, when the net continuous noise value exceeds a preset environmental noise threshold and the number of continuous windows reaches a preset continuous window number threshold, the present invention determines the automatic volume increment based on the difference between the net continuous noise value and the preset environmental noise threshold, and generates an automatic output volume based on the sum of the current output volume and the automatic volume increment. Thus, the device can increase the output volume when continuous environmental noise is present, ensuring that the playback content or voice prompts remain audible in a noisy kitchen environment; simultaneously, the automatic output volume does not exceed the device's maximum permissible output volume, preventing the output volume from exceeding the device's allowable range.

[0011] Furthermore, within a preset correction time after the automatic output volume is generated, when the knob operation indicates a decrease in volume and the near-field operation state indicates that the user has entered the operating distance of the knob device with a screen, a near-field mis-amplification event and a volume reduction amount are generated. Therefore, the user's behavior of adjusting the automatically increased volume in the opposite direction after approaching the knob device with a screen can be identified as a correction behavior for this automatic amplification, rather than being treated as a normal volume preference.

[0012] Finally, this invention generates a mis-adjustment scene fingerprint based on the effective noise window set corresponding to the near-field mis-amplification event and the volume reduction amount. When the effective noise window set matches the mis-adjustment scene fingerprint in subsequent events, the corresponding automatic volume increment is corrected based on the volume reduction amount, and then the current output volume is adjusted based on the corrected automatic volume increment. Therefore, when the same or similar continuous low-frequency noise scene reappears, the device can subtract the automatic volume increase amplitude that was previously corrected by the user, reducing the likelihood of the user hearing excessively high volumes when operating the knob at close range. Attached Figure Description

[0013] Figure 1 This is a flowchart illustrating an environmentally adaptive volume adjustment method for a rotary knob device with a screen, provided by an embodiment of the present invention. Detailed Implementation

[0014] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0015] like Figure 1As shown, an embodiment of the present invention proposes an environmentally adaptive volume adjustment method for a rotary control device with a screen. This method can be applied to rotary control devices with a display screen. The rotary control device with a screen may include a rotary operation component, a display screen, a microphone, a controller, and an audio output control module. The rotary operation component generates a rotary operation signal, the microphone collects environmental noise signals, and the controller processes the environmental noise signals and the rotary operation signal, adjusting the current output volume based on the processing result. The method includes: S1. Acquire the ambient noise signal, knob operation amount, current output volume, display wake-up status and near-field operation status of the screen-equipped rotary device, wherein the near-field operation status is used to characterize whether the user has entered the operation distance of the screen-equipped rotary device. S2. Continuously divide the environmental noise signal into sampling windows, calculate the average sound pressure level, peak sound pressure level, peak duration and low frequency energy ratio of each sampling window, and generate a sampling window feature set; S3. Based on the feature set of the sampling windows, sort the sampling windows by time. Mark the sampling windows whose difference between the peak sound pressure value and the average sound pressure value exceeds the preset impact difference threshold and whose peak duration is less than the preset impact duration threshold as transient impact windows, and remove the transient impact windows to generate an effective noise window set. S4. Based on the proportion of low-frequency energy in the effective noise window set, perform low-frequency screening and group them according to the continuity of sampling time to obtain the target continuous noise window group; generate the number of continuous windows and the net continuous noise value based on the number of sampling windows in the target continuous noise window group and the average sound pressure value of each sampling window. S5. When the net value of continuous noise exceeds the preset environmental noise threshold and the number of continuous windows reaches the preset number of continuous windows threshold, the automatic volume increment is determined based on the difference between the net value of continuous noise and the preset environmental noise threshold, and the automatic output volume is generated based on the sum of the current output volume and the automatic volume increment, wherein the automatic output volume does not exceed the maximum output volume allowed by the device. S6. Adjust the current output volume to the automatic output volume, and within the preset correction time after the automatic output volume is formed, when the knob operation indicates a decrease in volume and the near-field operation state is within the operation distance of the knob device with screen, generate a near-field false amplification event and a volume reduction amount. S7. Generate a mis-adjustment scene fingerprint based on the set of effective noise windows corresponding to near-field mis-amplification events and the amount of volume reduction. S8. When the subsequent set of effective noise windows matches the fingerprint of the misadjusted scene, the corresponding automatic volume increment is corrected according to the volume reduction amount, and the current output volume is adjusted according to the corrected automatic volume increment.

[0016] In this embodiment of the invention, the method acquires ambient noise signals, knob operation amounts, current output volume, display wake-up status, and near-field operation status, enabling the volume adjustment process to simultaneously incorporate both ambient noise and near-field operation judgments. The near-field operation status characterizes whether the user has entered the operating distance of the knob-equipped device with a screen, allowing subsequent volume-lowering operations to establish a correspondence with the near-field operation scenario, reducing interference from far-field noise changes or ordinary volume adjustments on the judgment of false amplification.

[0017] This method continuously divides the environmental noise signal into sampling windows and calculates the average sound pressure level, peak sound pressure level, peak duration, and low-frequency energy proportion, ensuring that each sampling window possesses sound pressure intensity, impact characteristics, and low-frequency characteristics. Based on the difference between the peak and average sound pressure levels and the peak duration, transient impact windows are eliminated. Instantaneous sounds such as spatula collisions, object strikes, and short, sharp sounds are not included in the subsequent continuous noise assessment, making the volume automatically increase less susceptible to triggering by short-duration impact sounds.

[0018] This method performs low-frequency screening of the effective noise window set based on the proportion of low-frequency energy, and groups the screened sampling windows according to the continuity of sampling time. This allows environmental noises with continuous low-frequency characteristics, such as those from range hoods and fans, to be identified as target continuous noise window groups. The number of continuous windows is generated based on the number of sampling windows in the target continuous noise window group, and a net continuous noise value is generated based on the average sound pressure level of each sampling window. This ensures that volume adjustment judgments are constrained by both noise intensity and duration, reducing misjudgments caused by fluctuations in individual sampling windows.

[0019] This method determines the automatic volume increment based on the difference between the net continuous noise value and the preset ambient noise threshold when the number of continuous windows reaches the preset continuous window number threshold. It then generates the automatic output volume based on the sum of the current output volume and the automatic volume increment. Therefore, the screen-equipped rotary device can increase the output volume in the presence of continuous ambient noise, ensuring that the playback content or prompts remain audible in that environment. Simultaneously, it limits the automatic output volume to the maximum allowable output volume of the device, preventing it from exceeding the device's permitted range.

[0020] This method identifies near-field amplification events as occurring within a preset correction time after the automatic output volume is generated, where the knob operation indicates a decrease in volume and the user is within the operating distance of the device with the screen. The method then determines the volume reduction amount based on the automatic output volume and the reduced volume. Therefore, when a user adjusts the automatically increased volume in the opposite direction after approaching the device with the screen, this operation is recorded as a correction result for the previous automatic amplification, rather than being simply treated as a normal volume preference.

[0021] This method generates a mis-adjustment scene fingerprint based on the effective noise window set corresponding to near-field mis-amplification events and the volume reduction amount, establishing a correlation between the noise scene corrected by the user and the corresponding volume reduction amount. When the subsequent effective noise window set matches the mis-adjustment scene fingerprint, the corresponding automatic volume increment is corrected according to the volume reduction amount, and the current output volume is adjusted according to the corrected automatic volume increment. This ensures that when the same or similar continuous low-frequency noise scene reappears, the device with a screen and knob can deduct the automatic volume increase amplitude that was previously corrected by the user, reducing the possibility of the user hearing excessively high output volume when operating at close range.

[0022] In one embodiment of the present invention, a preset impact difference threshold and a preset impact duration threshold are used to identify transient impact windows. The preset impact difference threshold is determined based on the distribution of the difference between the peak sound pressure level and the average sound pressure level in the transient impact noise sample and the continuous environmental noise sample; the preset impact duration threshold is determined based on the distribution of the peak sound pressure level duration in the transient impact noise sample and the continuous environmental noise sample. Specifically, the sounds of chopping vegetables, clattering spatulas, clattering tableware, or short knocking sounds are collected as transient impact noise samples, and the noise from a range hood, fan, or air conditioning duct is collected as continuous environmental noise samples; the difference between the peak sound pressure level and the average sound pressure level and the peak duration within the sampling window corresponding to each sample are calculated respectively, and the value boundaries that can distinguish between transient impact noise samples and continuous environmental noise samples are determined as the preset impact difference threshold and the preset impact duration threshold.

[0023] In one embodiment of the present invention, a preset environmental noise threshold is used to determine whether the net continuous noise level reaches the noise level requiring volume compensation. The preset environmental noise threshold can be determined based on the sound pressure level at which the output sound of the screen-equipped rotary device is masked by environmental noise in the target usage scenario. Specifically, when the screen-equipped rotary device is at different current output volumes, environmental noise samples are collected from the range hood, fan, or other continuous background noise sources, and the net continuous noise levels corresponding to when the user can hear the device's output sound normally and when it becomes difficult to hear the device's output sound are recorded. The preset environmental noise threshold is determined based on the boundaries of the aforementioned net continuous noise levels.

[0024] In one embodiment of the present invention, a preset duration window number threshold is used to determine whether continuous noise meets the condition of continuous existence in time. The preset duration window number threshold is determined based on the shortest duration that the continuous noise needs to be maintained and the time length corresponding to a single sampling window. Specifically, the shortest duration used to determine the existence of continuous noise is converted into the number of sampling windows, and the converted number of sampling windows is determined as the preset duration window number threshold. Thus, only noise window groups with continuous sampling time and a number that reaches the preset duration window number threshold are considered as continuous noise and participate in the generation and judgment of automatic volume increment.

[0025] In one embodiment of the present invention, a preset correction duration after the automatic output volume is formed is used to limit the time correlation between the user's reverse volume reduction operation and the current automatic output volume. This preset correction duration is determined based on the user's reaction time when performing a volume reduction operation after the automatic output volume is formed. Specifically, the time it takes for the user to approach the screen-equipped rotary device and reduce the volume after the automatic output volume is formed is collected, and the time range covering the user's reverse adjustment behavior for the current automatic output volume is determined as the preset correction duration. Volume reduction operations occurring within this preset correction duration are used to generate near-field amplification events and volume reduction amounts.

[0026] In a preferred embodiment of the present invention, the environmental noise signal is continuously divided into sampling windows, and the average sound pressure level, peak sound pressure level, peak duration, and low-frequency energy ratio of each sampling window are calculated to generate a sampling window feature set, including: The environmental noise digital sampling sequence collected by the microphone in the device with screen and knob is obtained, and the environmental noise digital sampling sequence is converted into a sound pressure sampling sequence according to the device's acoustic calibration coefficient; According to the preset sampling window duration and preset sliding step size, the sound pressure sampling sequence is continuously divided to generate multiple sampling windows arranged according to sampling time. Each sampling window is configured with a sampling window identifier and sampling time information, including the start sampling time and end sampling time of the sampling window. For any given sampling window, extract multiple sound pressure sampling values ​​within that window, sum the squares of the multiple sound pressure sampling values ​​and take the average value, then take the square root of the average value to generate the root mean square value of the sound pressure for that sampling window, and determine the root mean square value of the sound pressure as the average sound pressure value of that sampling window. For any sampling window, extract the absolute value of each sound pressure sampling value within the sampling window, and determine the sound pressure sampling value with the largest absolute value as the peak sampling value. Generate the peak sound pressure value of the sampling window based on the peak sampling value. For any sampling window, the sampling point corresponding to the peak sampling value is taken as the peak sampling point. Continuous sampling points that satisfy the peak preservation condition are searched forward and backward from the peak sampling point, respectively. The peak preservation condition is that the difference between the absolute value of the sound pressure corresponding to the current sampling point and the peak sound pressure value does not exceed a preset peak preservation difference. Based on the number of continuous sampling points that satisfy the peak preservation condition and the sampling frequency, the duration corresponding to the continuous sampling points is calculated, and this duration is determined as the peak duration of the sampling window. For any given sampling window, the sound pressure sampling values ​​within that window are deDC-processed, and then frequency-domain transformed to generate frequency-domain energy distribution data. Based on a preset low-frequency cutoff frequency, low-frequency band energy with frequencies not higher than the preset low-frequency cutoff frequency is extracted from the frequency-domain energy distribution data, and the total energy of the entire frequency band is calculated based on the frequency-domain energy distribution data. When the total energy of the entire frequency band is greater than zero, the ratio of low-frequency band energy to the total energy of the entire frequency band is determined as the low-frequency energy percentage. When the total energy of the entire frequency band is equal to zero, the low-frequency energy percentage is determined to be zero, or the corresponding sampling window is marked as an invalid sampling window. The sampling window identifier, sampling time information, average sound pressure level, peak sound pressure level, peak duration, and low-frequency energy ratio corresponding to the same sampling window are bound together to generate single-window feature data; the single-window feature data corresponding to multiple sampling windows are summarized in the order of sampling time to generate a sampling window feature set.

[0027] The average sound pressure level (SPL), peak SPL, and preset peak hold difference are measured in the same unit. The preset sampling window duration, preset sliding step size, preset peak hold difference, and preset low-frequency cutoff frequency can be obtained through factory calibration. Specifically, continuous low-frequency noise samples, short-term impact noise samples, and ordinary voice samples are collected in the target usage scenario. The preset sampling window duration and preset sliding step size are determined based on the noise change response time, sampling window characteristic stability, and device computing resources. The preset peak hold difference is determined based on the sound pressure drop distribution near the peak in the short-term impact noise samples. The preset low-frequency cutoff frequency is determined based on the frequency range of continuous low-frequency noise in the target usage scenario.

[0028] In one embodiment of the present invention, when the net value of continuous noise exceeds a preset environmental noise threshold and the number of continuous windows reaches a preset number threshold of continuous windows, the net value of continuous noise and the preset environmental noise threshold are represented by internal measurement values ​​corresponding to the device volume adjustment unit, and the difference between the net value of continuous noise and the preset environmental noise threshold is used as the automatic volume increment.

[0029] The current output volume is added to the automatic volume increment to generate a candidate automatic output volume. When the candidate automatic output volume does not exceed the maximum output volume allowed by the device, the candidate automatic output volume is determined as the automatic output volume. When the candidate automatic output volume exceeds the maximum output volume allowed by the device, the maximum output volume allowed by the device is determined as the automatic output volume.

[0030] When the net continuous noise value does not exceed the preset environmental noise threshold, or the number of continuous windows does not reach the preset continuous window number threshold, the current output volume remains unchanged.

[0031] In a preferred embodiment of the present invention, the process of generating near-field amplification events and volume reduction amounts includes: Within the preset correction time after the automatic output volume is generated, the direction of the knob operation is identified and the volume is converted according to the knob operation amount, and the volume adjustment direction and the adjusted volume are generated. When the volume adjustment direction is to decrease the volume, and the adjusted volume is lower than the automatic output volume, the adjusted volume will be determined as the decreased volume, and a reverse correction mark will be generated. The near-field operation status is read based on the reverse correction mark. When the near-field operation status indicates that the device has entered the operating distance of the screen-equipped rotary device, a near-field confirmation mark is generated. Based on the reverse correction flag and the near-field confirmation flag, generate a near-field misampling event corresponding to the current automatic output volume; When multiple reverse correction markers are generated within the same preset correction time, the reverse correction marker with the earliest formation time is selected as the trigger marker for the near-field false amplification event based on the formation time of the reverse correction marker. The volume reduction amount is determined based on the volume reduction after the automatic output volume and the corresponding trigger flag, and the formation time of the automatic output volume, the generation time of the near-field mis-amplification event, the trigger flag, and the volume reduction amount are written into the near-field mis-amplification event.

[0032] In this embodiment of the invention, the method identifies the direction of the knob operation and converts it into volume within a preset correction time after the automatic output volume is generated, and generates a reverse correction mark when the adjusted volume is lower than the automatic output volume. After generating a near-field confirmation mark in conjunction with the near-field operation status, the method can identify the volume reduction operation performed by the user when approaching the screen-equipped knob device as a correction behavior for the current automatic output volume. By selecting the earliest generated reverse correction mark as the trigger mark, it can avoid the repeated recording of multiple consecutive volume reduction operations within the same preset correction time. The volume reduction amount is determined based on the automatic output volume and the reduced volume, allowing the actual reduction magnitude of the automatic volume increase to be saved, providing a clear data source for subsequent recording of misadjustment scenarios.

[0033] In a preferred embodiment of the present invention, the process of generating a misadjustment scene fingerprint includes: Based on the formation time of the automatic output volume in the near-field false amplification event and the generation time of the near-field false amplification event, a sampling window located between the two times is extracted from the effective noise window set to generate an initial fingerprint extraction window set; The number of sampling windows in the initial fingerprint extraction window set is compared with the number of preset fingerprint windows. When the number of sampling windows is greater than or equal to the number of preset fingerprint windows, the initial fingerprint extraction window set is determined as the fingerprint extraction window set. When the number of sampling windows is less than the preset number of fingerprint windows, sampling windows are added from the set of effective noise windows adjacent to the formation time of the automatic output volume until the number of sampling windows is greater than or equal to the preset number of fingerprint windows, and a fingerprint extraction window set is generated. The average sound pressure value of each sampling window in the fingerprint extraction window set is sorted numerically to determine the lower limit and upper limit of the average sound pressure value interval, thus generating the average sound pressure value interval. The low-frequency energy ratio of each sampling window in the fingerprint extraction window set is numerically sorted to determine the lower limit and upper limit of the low-frequency energy ratio interval, thus generating the low-frequency energy ratio interval. Based on the number of sampling windows in the fingerprint extraction window set and the preset window number tolerance, a range of continuous window number is generated; The average sound pressure level range, low-frequency energy ratio range, duration window number range, and volume reduction amount are bound and stored to generate a fingerprint of the mis-tuning scenario.

[0034] In this embodiment of the invention, the method extracts sampling windows within a corresponding time range from the effective noise window set based on the automatic output volume formation time and the near-field amplification event generation time in the near-field amplification event. When the number of sampling windows is insufficient, sampling windows are supplemented from adjacent effective noise window sets, ensuring that the noise data used to generate the fingerprint of the misadjusted scene has a complete time segment. The noise intensity, low-frequency characteristics, and duration when corrected by the user are recorded through the average sound pressure level range, the low-frequency energy proportion range, and the duration window number range. These ranges are then bound to the volume reduction amount for storage, enabling the user's correction behavior to be associated with the corresponding noise scene, providing a matching basis for subsequent volume correction in similar noise scenes.

[0035] In one embodiment of the present invention, a preset number of fingerprint windows is used to ensure that the fingerprint in the mis-amplified scenario has a sufficient number of sampling window data. The preset number of fingerprint windows is determined based on the number of sampling windows required for the average sound pressure level range, the low-frequency energy proportion range, and the number of continuous windows range in the mis-amplified scenario fingerprint. Specifically, in the scenario where a user generates a near-field mis-amplified event, the number of effective noise windows that can be obtained between the automatic output volume formation time and the near-field mis-amplified event generation time is counted, and combined with the formation stability of the average sound pressure level range and the low-frequency energy proportion range, the minimum number of sampling windows used to generate the mis-amplified scenario fingerprint is determined, and this minimum number of sampling windows is determined as the preset number of fingerprint windows.

[0036] In one embodiment of the present invention, a preset window number tolerance is used to form the range of the number of persistent windows in the mis-tuned scenario fingerprint. The preset window number tolerance is determined based on the fluctuation range of the number of effective noise windows under the same or similar noise scenarios. Specifically, in the case of range hoods, fans, or other continuous low-frequency noise scenarios, fingerprint extraction window sets corresponding to near-field amplification events generated by the user are collected multiple times. The variation range of the number of sampling windows in each fingerprint extraction window set relative to the center value of the number of persistent windows is statistically analyzed, and the difference in the number of windows that can cover this variation range is determined as the preset window number tolerance. Thus, the number of persistent windows can fluctuate to a certain extent during subsequent matching of the mis-tuned scenario fingerprint.

[0037] In a preferred embodiment of the present invention, when the subsequent set of effective noise windows matches the fingerprint of the mis-adjusted scene, the corresponding automatic volume increment is corrected according to the volume reduction amount, and the current output volume is adjusted according to the corrected automatic volume increment, including: Based on the subsequent set of effective noise windows, the subsequent average sound pressure level range, the subsequent low-frequency energy ratio range, and the subsequent duration window number range are generated. The subsequent average sound pressure value range is compared with the average sound pressure value range in the mistuned scene fingerprint; the subsequent low-frequency energy proportion range is compared with the low-frequency energy proportion range in the mistuned scene fingerprint; and the subsequent continuous window number range is compared with the continuous window number range in the mistuned scene fingerprint. A fingerprint matching marker is generated when the subsequent average sound pressure value range falls within the average sound pressure value range of the mistuned scenario fingerprint, the subsequent low-frequency energy proportion range falls within the low-frequency energy proportion range of the mistuned scenario fingerprint, and the subsequent continuous window number range falls within the continuous window number range of the mistuned scenario fingerprint. Based on the fingerprint matching marker, the volume reduction amount is determined from the fingerprint of the mis-adjusted scene, and the volume reduction amount is subtracted from the automatic volume increment generated by the subsequent effective noise window set to generate candidate corrected volume increments; When the candidate corrected volume increment is not higher than zero, the corrected volume increment is set to zero, and the current output volume is kept from increasing with the automatic volume increment. When the candidate corrected volume increment is higher than zero, the candidate corrected volume increment is determined as the corrected volume increment, and the corrected output volume is generated based on the current output volume and the corrected volume increment; When the current output volume is adjusted to the corrected output volume, and a near-field mis-amplification event is generated again within the preset correction time after the corrected output volume is generated, the volume reduction amount in the regenerated near-field mis-amplification event is determined as the updated volume reduction amount and written into the mis-adjustment scene fingerprint.

[0038] In this embodiment of the invention, the method generates a subsequent average sound pressure level range, a subsequent low-frequency energy ratio range, and a subsequent duration window number range based on the subsequent effective noise window set. These ranges are then compared with the corresponding ranges in the mis-tuned scene fingerprint, ensuring that subsequent noise scenes are only corrected when the noise intensity, low-frequency characteristics, and duration all meet the matching conditions. After matching, the volume reduction amount is determined from the mis-tuned scene fingerprint, and this volume reduction amount is subtracted from the automatic volume increment generated based on the subsequent effective noise window set to generate a corrected volume increment. The corrected volume increment does not increase with the automatic volume increment when it is not higher than zero; only when the candidate corrected volume increment is higher than zero is the corrected output volume generated. This reduces the likelihood of excessively high automatic volume when similar noise scenes reappear.

[0039] In a preferred embodiment of the present invention, if the lower limit of the subsequent average sound pressure level interval is not lower than the lower limit of the average sound pressure level interval in the mistuned scenario fingerprint, and the upper limit of the subsequent average sound pressure level interval is not higher than the upper limit of the average sound pressure level interval in the mistuned scenario fingerprint, then the subsequent average sound pressure level interval is determined to fall within the average sound pressure level interval in the mistuned scenario fingerprint. The same applies to the low-frequency energy proportion interval and the duration window number interval.

[0040] In one embodiment of the present invention, the volume reduction in the misadjustment scenario fingerprint is not calculated separately from the average sound pressure level range, the low-frequency energy ratio range, or the number of continuous windows range. Instead, it is the actual correction amount formed after the user operates the screen-equipped knob device and reduces the volume within a preset correction time after the automatic output volume is formed. The volume reduction amount is used to characterize the magnitude of the reverse adjustment made by the user for this automatic output volume.

[0041] The mis-amplification scenario fingerprint is used to limit the reuse range of the volume reduction amount. Specifically, the mis-amplification scenario fingerprint binds and stores the average sound pressure level range, low-frequency energy ratio range, duration window number range, and volume reduction amount corresponding to the near-field mis-amplification event generated by the user. After the subsequently collected environmental noise signal is processed to form a subsequent set of effective noise windows, the volume reduction amount in the mis-amplification scenario fingerprint is only read when the average sound pressure level range, low-frequency energy ratio range, and duration window number range corresponding to the subsequent set of effective noise windows all match the corresponding ranges in the mis-amplification scenario fingerprint, and this volume reduction amount is used to correct the corresponding automatic volume increment.

[0042] Therefore, the volume reduction amount is not universally applicable to all subsequent noise scenarios. Instead, it is used only when the subsequent noise scenario matches the historical correction noise scenario represented by the misadjustment scenario fingerprint. In this way, the actual correction amount formed by the user in historical near-field operations can be limited to reuse in the same or similar noise scenarios, avoiding the direct application of a single volume reduction operation to irrelevant noise scenarios.

[0043] After correcting the automatic volume increment based on the volume reduction and generating the corrected output volume, if the user reduces the volume again within the preset correction time after the corrected output volume is generated, and a near-field mis-amplification event is generated again, the volume reduction amount stored in the mis-tuning scenario fingerprint is updated according to the volume reduction amount in the regenerated near-field mis-amplification event. The updated volume reduction amount continues to be stored corresponding to the average sound pressure level range, low-frequency energy ratio range, and duration window number range in the mis-tuning scenario fingerprint, and is used to correct the automatic volume increment when the effective noise window set is matched again in the future.

[0044] Through the above process, the misadjusted scenario fingerprint forms a closed-loop processing process: the volume reduction is generated by the user's near-field correction, the volume reduction is reused by limiting the noise range, the volume increment is corrected by subsequent matching, and the volume reduction is updated by the correction again.

[0045] In a preferred embodiment of the present invention, the process of generating the range of the number of persistent windows includes: The number of sampling windows in the fingerprint extraction window set is determined as the center value of the number of continuous windows; The lower limit of the continuous window number range is generated by extending the center value of the continuous window number downwards by the preset window number tolerance. Extend the center value of the number of persistent windows upwards by the preset window number tolerance to generate the upper limit of the range of the number of persistent windows; When the lower limit of the interval for the number of continuous windows is less than the number of a sampling window, the lower limit of the interval for the number of continuous windows is determined to be the number of a sampling window; Bind the lower limit of the duration window number range, the center value of the duration window number, and the upper limit of the duration window number range to generate the duration window number range; The range of continuous window quantity is stored in correspondence with the range of average sound pressure level, the range of low-frequency energy proportion, and the amount of volume reduction, and is used as the data items of the misadjustment scenario fingerprint.

[0046] In this embodiment of the invention, the method uses the number of sampling windows in the fingerprint extraction window set as the center value of the number of persistent windows, and generates a lower limit and an upper limit of the number of persistent windows interval based on a preset window number tolerance, so that the duration feature in the fingerprint of the mistuned scene has a matching range. By limiting the lower limit of the number of persistent windows interval to no less than the number of sampling windows, invalid boundaries in the number of persistent windows interval can be avoided. After storing the number of persistent windows interval in correspondence with the average sound pressure level interval, the low-frequency energy ratio interval, and the volume reduction amount, subsequent noise scene matching not only relies on sound pressure and low-frequency features, but also combines the noise duration, reducing the occurrence of short noise segments being treated as mistuned scenes.

[0047] In a preferred embodiment of the present invention, the process of forming the near-field operating state includes: Acquire the knob touch signal, knob press signal, knob operation time, and display wake-up status of the device with a screen and knob; When the knob touch signal is triggered or the knob press signal is triggered, a near-field candidate mark is generated; When the display is in the wake-up state, and the knob operation time is after the near-field candidate mark formation time, and the time difference between the two does not exceed the preset near-field confirmation time, a near-field confirmation result is generated. Based on the near-field confirmation results, the near-field operation status is determined to be the operating distance of the device with a screen and knob; When no near-field candidate marker is generated, or the display screen is not awake, or the knob operation time is not after the near-field candidate marker formation time, or the time difference between the knob operation time and the near-field candidate marker formation time exceeds the preset near-field confirmation time, the near-field operation state is determined as not entering the operation distance of the knob device with screen.

[0048] In this embodiment of the invention, the method establishes a near-field operation state through knob touch signals, knob press signals, knob operation amount formation time, and display screen wake-up state, allowing the device's own interaction signals to determine whether the user has entered the operating distance of the knob-equipped device. A near-field candidate marker is generated after the knob touch signal or knob press signal is triggered, and this marker is then confirmed by combining the display screen wake-up state and the knob operation amount formation time. This distinguishes between situations where the user is operating close to the device and situations involving changes in far-field environmental noise. If the touch, press, display wake-up, or time sequence conditions are not met, the near-field operation state is determined as not having entered the operating distance, reducing the possibility of irrelevant operations being used to generate near-field false amplification events.

[0049] In a preferred embodiment of the present invention, the preset near-field confirmation duration is determined based on the time interval between the user touching or pressing the knob to generate a knob operation. Specifically, the time interval between the user touching or pressing the knob and continuing to rotate, slide, or press to generate a knob operation is collected, and the preset near-field confirmation duration is determined based on this time interval. Thus, when the knob operation time is after the near-field candidate marker formation time, and the time difference between the two does not exceed the preset near-field confirmation duration, it can be confirmed that the user is within the operating distance of the knob device with a screen.

[0050] In a preferred embodiment of the present invention, the process of generating the net continuous noise value and the number of continuous windows includes: Extract the average sound pressure level, low-frequency energy percentage, and sampling time of each sampling window from the set of effective noise windows; Based on the low-frequency energy ratio of each sampling window, low-frequency screening is performed, and sampling windows whose low-frequency energy ratio reaches the preset low-frequency ratio threshold are determined as candidate continuous noise windows, generating a set of candidate continuous noise windows. Based on the sampling time continuity of each candidate persistent noise window in the candidate persistent noise window set, candidate persistent noise windows with continuous sampling time are grouped into the same candidate persistent noise window group. The candidate persistent noise window group is selected from the candidate persistent noise window group. The number of sampling windows reaches the preset persistent window number threshold, and the sampling time of the last candidate persistent noise window in the group is closest to the current time. This group is then determined as the target persistent noise window group. The number of persistent windows is generated based on the number of sampling windows in the target persistent noise window group; Extract the average sound pressure value of each sampling window in the target continuous noise window group to generate the target sound pressure sequence; The target sound pressure sequence is arranged in ascending order of numerical value. When the number of average sound pressure values ​​in the target sound pressure sequence is odd, the average sound pressure value in the middle position is determined as the net continuous noise value. When the number of average sound pressure values ​​in the target sound pressure sequence is even, the arithmetic mean of the two average sound pressure values ​​in the two middle positions is determined as the net continuous noise value.

[0051] In this embodiment of the invention, the method extracts the average sound pressure level, low-frequency energy ratio, and sampling time from the effective noise window set, and filters candidate persistent noise windows based on the low-frequency energy ratio, enabling persistent noise judgment to focus on environmental noise with low-frequency characteristics. Candidate persistent noise windows are grouped based on the continuity of sampling time, and the candidate persistent noise window group that meets the persistent window quantity threshold and is closest to the current time is selected as the target persistent noise window group, ensuring that the number of persistent windows reflects the duration of the current persistent noise. A persistent noise net value is generated based on the median of the average sound pressure level within the target persistent noise window group, preventing sound pressure fluctuations in a single sampling window from directly determining the automatic volume adjustment judgment.

[0052] In one embodiment of the present invention, a preset low-frequency proportion threshold is used to filter candidate continuous noise windows with low-frequency characteristics from the effective noise window set. The preset low-frequency proportion threshold is determined based on the distribution of low-frequency energy proportion between continuous low-frequency noise samples and non-continuous interference noise samples. Specifically, range hood noise, fan noise, and air conditioner duct noise are collected as continuous low-frequency noise samples, and human voices, prompts, chopping sounds, and spatula collision sounds are collected as non-continuous interference noise samples. The low-frequency energy proportion of each of the above samples within the sampling window is calculated, and the low-frequency energy proportion boundary that can distinguish between continuous low-frequency noise samples and non-continuous interference noise samples is determined as the preset low-frequency proportion threshold. Sampling windows that reach the preset low-frequency proportion threshold are determined as candidate continuous noise windows.

[0053] Embodiments of the present invention also provide an environmentally adaptive volume control system for a device with a screen and knob, the system comprising: The data acquisition module is used to acquire the environmental noise signal, knob operation amount, current output volume, display wake-up status and near-field operation status of the screen-equipped rotary device, wherein the near-field operation status is used to characterize whether the user has entered the operation distance of the screen-equipped rotary device. The sampling window feature generation module is used to continuously divide the environmental noise signal into sampling windows, calculate the average sound pressure value, peak sound pressure value, peak duration and low frequency energy ratio of each sampling window, and generate a sampling window feature set. The transient impact rejection module is used to sort the sampling windows by time according to the sampling window feature set, mark the sampling windows whose peak sound pressure value and average sound pressure value difference exceeds the preset impact difference threshold and whose peak duration is less than the preset impact duration threshold as transient impact windows, and reject the transient impact windows to generate an effective noise window set. The continuous noise generation module is used to perform low-frequency filtering based on the proportion of low-frequency energy in the effective noise window set, and to group them according to the continuity of sampling time to obtain the target continuous noise window group; based on the number of sampling windows in the target continuous noise window group and the average sound pressure value of each sampling window, the number of continuous windows and the net continuous noise value are generated. An automatic volume generation module is used to determine an automatic volume increment based on the difference between the net continuous noise value and the preset environmental noise threshold when the net continuous noise value exceeds a preset environmental noise threshold and the number of continuous windows reaches a preset continuous window number threshold, and to generate an automatic output volume based on the sum of the current output volume and the automatic volume increment, wherein the automatic output volume does not exceed the maximum output volume allowed by the device. The near-field false amplification event generation module is used to adjust the current output volume to the automatic output volume, and within the preset correction time after the automatic output volume is formed, when the knob operation indicates a decrease in volume and the near-field operation state is within the operating distance of the knob device with screen, a near-field false amplification event and a volume reduction amount are generated. The mis-amplification scene fingerprint generation module is used to generate a mis-amplification scene fingerprint based on the effective noise window set and volume reduction amount corresponding to the near-field mis-amplification event. The volume adjustment module is used to correct the corresponding automatic volume increment based on the volume reduction amount when the subsequent effective noise window set matches the fingerprint of the misadjusted scene, and adjust the current output volume based on the corrected automatic volume increment.

[0054] It should be noted that this system is a system corresponding to the above method. All implementation methods in the above method embodiments are applicable to this embodiment and can achieve the same technical effect.

[0055] Embodiments of the present invention also provide a computing device, including: a processor and a memory storing a computer program, wherein the computer program, when executed by the processor, performs the method described above. All implementations in the above method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0056] Embodiments of the present invention also provide a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the method described above. All implementations in the above method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0057] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. An environmentally adaptive volume adjustment method for a device with a screen and knob, characterized in that, The method includes: Acquire the ambient noise signal, knob operation amount, current output volume, display wake-up status and near-field operation status of the device with screen and knob, wherein the near-field operation status is used to characterize whether the user has entered the operation distance of the device with screen and knob. The environmental noise signal is continuously divided into sampling windows. The average sound pressure level, peak sound pressure level, peak duration and low frequency energy ratio of each sampling window are calculated to generate a sampling window feature set. Based on the feature set of the sampling windows, the sampling windows are sorted by time. Sampling windows whose difference between peak sound pressure value and average sound pressure value exceeds a preset impact difference threshold and whose peak duration is less than a preset impact duration threshold are marked as transient impact windows. Transient impact windows are then removed to generate an effective noise window set. Low-frequency filtering is performed based on the proportion of low-frequency energy in the effective noise window set, and the groups are formed by sampling time continuity to obtain the target continuous noise window group; the number of continuous windows and the net continuous noise value are generated based on the number of sampling windows in the target continuous noise window group and the average sound pressure value of each sampling window. When the net continuous noise value exceeds the preset environmental noise threshold and the number of continuous windows reaches the preset continuous window number threshold, the automatic volume increment is determined based on the difference between the net continuous noise value and the preset environmental noise threshold, and the automatic output volume is generated based on the sum of the current output volume and the automatic volume increment, wherein the automatic output volume does not exceed the maximum output volume allowed by the device. Adjust the current output volume to the automatic output volume, and within the preset correction time after the automatic output volume is formed, when the knob operation indicates a decrease in volume and the near-field operation state is within the operation distance of the knob device with screen, generate a near-field false amplification event and a volume reduction amount. Generate a mis-adjustment scene fingerprint based on the set of effective noise windows corresponding to near-field mis-amplification events and the amount of volume reduction. When the subsequent set of effective noise windows matches the fingerprint of the misadjusted scene, the corresponding automatic volume increment is corrected according to the volume reduction amount, and the current output volume is adjusted according to the corrected automatic volume increment.

2. The environmental adaptive volume adjustment method for a rotary device with a screen according to claim 1, characterized in that, The process of generating near-field amplification events and volume reduction amounts includes: Within the preset correction time after the automatic output volume is generated, the direction of the knob operation is identified and the volume is converted according to the knob operation amount, and the volume adjustment direction and the adjusted volume are generated. When the volume adjustment direction is to decrease the volume, and the adjusted volume is lower than the automatic output volume, the adjusted volume will be determined as the decreased volume, and a reverse correction mark will be generated. The near-field operation status is read based on the reverse correction mark. When the near-field operation status indicates that the device has entered the operating distance of the screen-equipped rotary device, a near-field confirmation mark is generated. Based on the reverse correction flag and the near-field confirmation flag, generate a near-field misampling event corresponding to the current automatic output volume; When multiple reverse correction markers are generated within the same preset correction time, the reverse correction marker with the earliest formation time is selected as the trigger marker for the near-field false amplification event based on the formation time of the reverse correction marker. The volume reduction amount is determined based on the volume reduction after the automatic output volume and the corresponding trigger flag, and the formation time of the automatic output volume, the generation time of the near-field mis-amplification event, the trigger flag, and the volume reduction amount are written into the near-field mis-amplification event.

3. The environmental adaptive volume adjustment method for a device with a screen and knob according to claim 2, characterized in that, The process of generating a misadjustment scenario fingerprint includes: Based on the formation time of the automatic output volume in the near-field false amplification event and the generation time of the near-field false amplification event, a sampling window located between the two times is extracted from the effective noise window set to generate an initial fingerprint extraction window set; The number of sampling windows in the initial fingerprint extraction window set is compared with the number of preset fingerprint windows. When the number of sampling windows is greater than or equal to the number of preset fingerprint windows, the initial fingerprint extraction window set is determined as the fingerprint extraction window set. When the number of sampling windows is less than the preset number of fingerprint windows, sampling windows are added from the set of effective noise windows adjacent to the formation time of the automatic output volume until the number of sampling windows is greater than or equal to the preset number of fingerprint windows, and a fingerprint extraction window set is generated. The average sound pressure value of each sampling window in the fingerprint extraction window set is sorted numerically to determine the lower limit and upper limit of the average sound pressure value interval, thus generating the average sound pressure value interval. The low-frequency energy ratio of each sampling window in the fingerprint extraction window set is numerically sorted to determine the lower limit and upper limit of the low-frequency energy ratio interval, thus generating the low-frequency energy ratio interval. Based on the number of sampling windows in the fingerprint extraction window set and the preset window number tolerance, a range of continuous window number is generated; The average sound pressure level range, low-frequency energy ratio range, duration window number range, and volume reduction amount are bound and stored to generate a fingerprint of the mis-tuning scenario.

4. The environmental adaptive volume adjustment method for a rotary device with a screen according to claim 1, characterized in that, When the subsequent set of effective noise windows matches the fingerprint of the mis-adjusted scene, the corresponding automatic volume increment is corrected according to the volume reduction amount, and the current output volume is adjusted according to the corrected automatic volume increment, including: Based on the subsequent set of effective noise windows, the subsequent average sound pressure level range, the subsequent low-frequency energy ratio range, and the subsequent duration window number range are generated. The subsequent average sound pressure value range is compared with the average sound pressure value range in the mistuned scene fingerprint; the subsequent low-frequency energy proportion range is compared with the low-frequency energy proportion range in the mistuned scene fingerprint; and the subsequent continuous window number range is compared with the continuous window number range in the mistuned scene fingerprint. A fingerprint matching marker is generated when the subsequent average sound pressure value range falls within the average sound pressure value range of the mistuned scenario fingerprint, the subsequent low-frequency energy proportion range falls within the low-frequency energy proportion range of the mistuned scenario fingerprint, and the subsequent continuous window number range falls within the continuous window number range of the mistuned scenario fingerprint. Based on the fingerprint matching marker, the volume reduction amount is determined from the fingerprint of the mis-adjusted scene, and the volume reduction amount is subtracted from the automatic volume increment generated by the subsequent effective noise window set to generate candidate corrected volume increments; When the candidate corrected volume increment is not higher than zero, the corrected volume increment is set to zero, and the current output volume is kept from increasing with the automatic volume increment. When the candidate corrected volume increment is higher than zero, the candidate corrected volume increment is determined as the corrected volume increment, and the corrected output volume is generated based on the current output volume and the corrected volume increment; When the current output volume is adjusted to the corrected output volume, and a near-field mis-amplification event is generated again within the preset correction time after the corrected output volume is generated, the volume reduction amount in the regenerated near-field mis-amplification event is determined as the updated volume reduction amount and written into the mis-adjustment scene fingerprint.

5. The environmental adaptive volume adjustment method for a rotary device with a screen according to claim 3, characterized in that, The process of generating the range of duration windows includes: The number of sampling windows in the fingerprint extraction window set is determined as the center value of the number of continuous windows; The lower limit of the continuous window number range is generated by extending the center value of the continuous window number downwards by the preset window number tolerance. Extend the center value of the number of persistent windows upwards by the preset window number tolerance to generate the upper limit of the range of the number of persistent windows; When the lower limit of the interval for the number of continuous windows is less than the number of a sampling window, the lower limit of the interval for the number of continuous windows is determined to be the number of a sampling window; Bind the lower limit of the duration window number range, the center value of the duration window number, and the upper limit of the duration window number range to generate the duration window number range; The range of continuous window quantity is stored in correspondence with the range of average sound pressure level, the range of low-frequency energy proportion, and the amount of volume reduction, and is used as the data items of the misadjustment scenario fingerprint.

6. The environmental adaptive volume adjustment method for a rotary device with a screen according to claim 1, characterized in that, The formation process of the near-field operating state includes: Acquire the knob touch signal, knob press signal, knob operation time, and display wake-up status of the device with a screen and knob; When the knob touch signal is triggered or the knob press signal is triggered, a near-field candidate mark is generated; When the display is in the wake-up state, and the knob operation time is after the near-field candidate mark formation time, and the time difference between the two does not exceed the preset near-field confirmation time, a near-field confirmation result is generated. Based on the near-field confirmation results, the near-field operation status is determined to be the operating distance of the device with a screen and knob; When no near-field candidate marker is generated, or the display screen is not awake, or the knob operation time is not after the near-field candidate marker formation time, or the time difference between the knob operation time and the near-field candidate marker formation time exceeds the preset near-field confirmation time, the near-field operation state is determined as not entering the operation distance of the knob device with screen.

7. The environmental adaptive volume adjustment method for a rotary device with a screen according to claim 1, characterized in that, The process of generating the net continuous noise value and the number of continuous windows includes: Extract the average sound pressure level, low-frequency energy percentage, and sampling time of each sampling window from the set of effective noise windows; Based on the low-frequency energy ratio of each sampling window, low-frequency screening is performed, and sampling windows whose low-frequency energy ratio reaches the preset low-frequency ratio threshold are determined as candidate continuous noise windows, generating a set of candidate continuous noise windows. Based on the sampling time continuity of each candidate persistent noise window in the candidate persistent noise window set, candidate persistent noise windows with continuous sampling time are grouped into the same candidate persistent noise window group. The candidate persistent noise window group is selected from the candidate persistent noise window group. The number of sampling windows reaches the preset persistent window number threshold, and the sampling time of the last candidate persistent noise window in the group is closest to the current time. This group is then determined as the target persistent noise window group. The number of persistent windows is generated based on the number of sampling windows in the target persistent noise window group; Extract the average sound pressure value of each sampling window in the target continuous noise window group to generate the target sound pressure sequence; The target sound pressure sequence is arranged in ascending order of numerical value. When the number of average sound pressure values ​​in the target sound pressure sequence is odd, the average sound pressure value in the middle position is determined as the net continuous noise value. When the number of average sound pressure values ​​in the target sound pressure sequence is even, the arithmetic mean of the two average sound pressure values ​​in the two middle positions is determined as the net continuous noise value.

8. An environmentally adaptive volume control system for a device with a screen and knob, characterized in that, The system, used in any one of claims 1 to 7, comprises: The data acquisition module is used to acquire the environmental noise signal, knob operation amount, current output volume, display wake-up status and near-field operation status of the screen-equipped rotary device, wherein the near-field operation status is used to characterize whether the user has entered the operation distance of the screen-equipped rotary device. The sampling window feature generation module is used to continuously divide the environmental noise signal into sampling windows, calculate the average sound pressure value, peak sound pressure value, peak duration and low frequency energy ratio of each sampling window, and generate a sampling window feature set. The transient impact rejection module is used to sort the sampling windows by time according to the sampling window feature set, mark the sampling windows whose peak sound pressure value and average sound pressure value difference exceeds the preset impact difference threshold and whose peak duration is less than the preset impact duration threshold as transient impact windows, and reject the transient impact windows to generate an effective noise window set. The continuous noise generation module is used to perform low-frequency filtering based on the proportion of low-frequency energy in the effective noise window set, and to group them according to the continuity of sampling time to obtain the target continuous noise window group; based on the number of sampling windows in the target continuous noise window group and the average sound pressure value of each sampling window, the number of continuous windows and the net continuous noise value are generated. An automatic volume generation module is used to determine an automatic volume increment based on the difference between the net continuous noise value and the preset environmental noise threshold when the net continuous noise value exceeds a preset environmental noise threshold and the number of continuous windows reaches a preset continuous window number threshold, and to generate an automatic output volume based on the sum of the current output volume and the automatic volume increment, wherein the automatic output volume does not exceed the maximum output volume allowed by the device. The near-field false amplification event generation module is used to adjust the current output volume to the automatic output volume, and within the preset correction time after the automatic output volume is formed, when the knob operation indicates a decrease in volume and the near-field operation state is within the operating distance of the knob device with screen, a near-field false amplification event and a volume reduction amount are generated. The mis-amplification scene fingerprint generation module is used to generate a mis-amplification scene fingerprint based on the effective noise window set and volume reduction amount corresponding to the near-field mis-amplification event. The volume adjustment module is used to correct the corresponding automatic volume increment based on the volume reduction amount when the subsequent effective noise window set matches the fingerprint of the misadjusted scene, and adjust the current output volume based on the corrected automatic volume increment.

9. A computing device, characterized in that, include: One or more processors; A storage device for storing one or more programs, which, when executed by one or more processors, cause the one or more processors to implement the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a program that, when executed by a processor, implements the method as described in any one of claims 1 to 7.

Images