Sound box volume intelligent adjustment method and device combining environmental noise monitoring and voice recognition

Abstract

The present application relates to a sound box volume intelligent adjustment method combining environmental noise monitoring and voice recognition, comprising the following: obtaining a first decibel value and a second decibel value of adjacent sampling intervals without human voice interference respectively; calculating the difference between the second decibel value and the first decibel value to obtain a decibel change value; judging the environmental noise change based on the decibel change value, if the environmental noise change is larger or smaller, triggering the voice recognition function, obtaining the feedback voice of the user and based on this to adaptively adjust the sound box volume. The present application uses the difference between the second decibel value and the first decibel value without human voice interference to approximately estimate the environmental noise change, when the environmental noise changes, the voice recognition function is triggered, the voice interaction with the user is carried out, and the sound box volume is adaptively adjusted, on the one hand, the volume can be adjusted in time when the noise fluctuates to adapt to the user's listening to the sound box content, on the other hand, the interaction with the user increases the user's experience.

Description

Technical Field

[0001] This invention relates to the field of smart speaker technology, and in particular to a method and apparatus for intelligent volume adjustment of a speaker that combines environmental noise monitoring and speech recognition. Background Technology

[0002] A speaker is a device that converts audio signals into sound. With technological advancements, many intelligent products have emerged in the speaker industry. Among these, intelligent volume control is a popular feature. Current technologies often employ noise balancing algorithms to output noise-resistant sound when real-time ambient noise changes are detected. While this method achieves noise balancing, it is computationally intensive and costly. In some applications, such as outdoor or in-vehicle use, users may simply need the speaker to adjust the volume appropriately to a suitable level without interfering with listening when ambient noise fluctuations exceed a certain threshold. In such scenarios, intelligent volume adjustment in response to ambient noise fluctuations is a key technological challenge. Summary of the Invention

[0003] The purpose of this invention is to at least address one of the shortcomings of the prior art by providing a method for intelligent adjustment of speaker volume that combines environmental noise monitoring and speech recognition.

[0004] To achieve the above objectives, the present invention adopts the following technical solution: Specifically, a method for intelligent speaker volume adjustment combining environmental noise monitoring and speech recognition is proposed, including the following: The system acquires the first sound signal and the first decibel value of the target speaker during operation, determines whether user voice exists in the sound signal, discards the first decibel value if it exists, and retains the first decibel value if it does not exist, until the first decibel value is retained. After a preset duration, the second sound signal and the second decibel value of the target speaker during operation are acquired, and it is determined whether there is user voice in the sound signal. If there is, the second decibel value is discarded; if not, the second decibel value is retained until the second decibel value is retained. The difference between the second decibel value and the first decibel value is used to obtain the decibel change value. Based on the decibel change value, the environmental noise change is determined. If the environmental noise changes to increase or decrease, the voice recognition function is triggered to obtain the user's feedback voice and adjust the speaker volume accordingly.

[0005] Furthermore, specifically, determining whether user voice exists in the sound signal includes, The audio signal is passed through an FIR low-pass filter to remove high-frequency aliasing noise. Then, gain calibration is performed to convert the electrical signal amplitude into a sound pressure-related uniform scale. After that, the VAD algorithm is used to determine whether the calculated frame energy and spectral flatness meet the corresponding judgment rules. If they meet the rules, it is determined that there is user speech in the audio signal. If they do not meet the rules, it is determined that there is no user speech in the audio signal.

[0006] Furthermore, specifically, judging the change in environmental noise based on the aforementioned decibel change value includes, A reasonable range for decibel changes is preset. If the decibel change exceeds the maximum value of the reasonable range, the environmental noise change is determined to be increasing. If the decibel change exceeds the minimum value of the reasonable range, the environmental noise change is determined to be decreasing.

[0007] Furthermore, specifically, the process of triggering the voice recognition function, obtaining the user's feedback voice, and adaptively adjusting the speaker volume based on this includes: If the ambient noise changes and increases, the user is asked whether to adjust the speaker volume. If the user confirms the adjustment, the system uses (a, b) as the adaptive volume adjustment range to interact with the user via voice and thus adaptively increase the speaker volume. Here, a is the current volume and b is the preset maximum volume. If the ambient noise decreases, the user is asked whether to adjust the speaker volume. If the user confirms the adjustment, the system uses [c, a) as the adaptive volume adjustment range and interacts with the user via voice to adaptively reduce the speaker volume. c is the preset minimum volume value.

[0008] Furthermore, specifically, the process of using (a, b) as the volume adaptive adjustment range to interact with the user via voice and thereby adaptively increase the speaker volume includes: A volume value x is randomly selected from (a, b) as the initial volume adjustment, and user voice feedback is obtained. The user voice feedback is then subjected to speech recognition. If the speech recognition result is confirmation, the initial volume adjustment is used as the final volume. If the speech recognition result indicates that the volume needs to be increased, (x, b) is used as the new adaptive volume adjustment range, and the above steps are performed. If the speech recognition result indicates that the volume needs to be decreased, (a, x) is used as the new adaptive volume adjustment range, and the above steps are performed. If the speech recognition result indicates a vague intention, [xy, x) ∩ (x, x+y) is used as the new adaptive volume adjustment range, and the above steps are performed. y is a preset adjustment parameter, until the speech recognition result is confirmation.

[0009] Furthermore, specifically, the process of using [c, a) as the volume adaptive adjustment range to interact with the user via voice and thereby adaptively reduce the speaker volume includes, A volume value p is randomly selected from [c, a) as the initial volume adjustment, and user voice feedback is obtained. The user voice feedback is then subjected to speech recognition. If the speech recognition result is confirmation, the initial volume adjustment is used as the final volume. If the speech recognition result indicates that the volume needs to be increased, (p, a) is used as the new adaptive volume adjustment range, and the above steps are performed. If the speech recognition result indicates that the volume needs to be decreased, [c, p) is used as the new adaptive volume adjustment range, and the above steps are performed. If the speech recognition result indicates a vague intention, [pq, p) ∩ (p, p+q] is used as the new adaptive volume adjustment range, and the above steps are performed, where q is a preset adjustment parameter, until the speech recognition result is confirmation.

[0010] Furthermore, specifically, the process of obtaining speech recognition results includes: A first dictionary, a second dictionary, and a third dictionary are pre-built. When the user's speech content obtained from speech recognition matches the first dictionary, the speech recognition result is determined to be "confirmed". When the user's speech content obtained from speech recognition matches the second dictionary, the speech recognition result is determined to be "needs to increase". When the user's speech content obtained from speech recognition matches the third dictionary, the speech recognition result is determined to be "needs to decrease". If the user's speech content obtained from speech recognition does not match any word in the above three dictionaries, the speech recognition result is determined to be "fuzzy intention".

[0011] Furthermore, specifically, the FIR low-pass filter used is a 64th order filter with a cutoff frequency of 22kHz.

[0012] This invention also proposes a smart speaker volume adjustment device that combines environmental noise monitoring and voice recognition, comprising the following: The first data acquisition module is used to acquire the first sound signal and the first decibel value when the target speaker is running, determine whether there is user voice in the sound signal, discard the first decibel value if it exists, and retain the first decibel value if it does not exist, until the first decibel value is retained. The second data acquisition module is used to acquire the second sound signal and the second decibel value of the target speaker after a preset time, determine whether there is user voice in the sound signal, discard the second decibel value if it exists, and retain the second decibel value if it does not exist, until the second decibel value is retained. The difference calculation module is used to calculate the difference between the second decibel value and the first decibel value to obtain the decibel change value. The volume adjustment module is used to determine the change in ambient noise based on the decibel change value. If the ambient noise changes to increase or decrease, the voice recognition function is triggered to obtain the user's feedback voice and adjust the speaker volume adaptively accordingly.

[0013] The beneficial effects of this invention are as follows: This invention proposes a method and device for intelligent speaker volume adjustment that combines environmental noise monitoring and speech recognition. By acquiring a first decibel value and a second decibel value with no human noise interference at adjacent sampling intervals, and using the difference between the second decibel value and the first decibel value to approximate the change in environmental noise, the speech recognition function is triggered when the environmental noise changes, and the speaker volume is adaptively adjusted by interacting with the user. On the one hand, the volume can be adjusted in a timely manner to suit the user's listening experience when the noise fluctuates; on the other hand, the interaction with the user enhances the user experience. Attached Figure Description

[0014] The above and other features of this disclosure will become more apparent from the detailed description of the embodiments illustrated in conjunction with the accompanying drawings. In the accompanying drawings, the same reference numerals denote the same or similar elements. Obviously, the drawings described below are merely some embodiments of this disclosure. For those skilled in the art, other drawings can be obtained from these drawings without any creative effort. In the drawings: Figure 1 The flowchart shown is a process for the intelligent speaker volume adjustment method combining environmental noise monitoring and speech recognition according to the present invention. Detailed Implementation

[0015] The following will provide a clear and complete description of the concept, specific structure, and technical effects of the present invention in conjunction with embodiments and accompanying drawings, so as to fully understand the purpose, solution, and effects of the present invention. It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. The same reference numerals used throughout the accompanying drawings indicate the same or similar parts.

[0016] Example 1, referring to Figure 1 This invention proposes a method for intelligent adjustment of speaker volume that combines environmental noise monitoring and speech recognition, including the following: Step 110: Obtain the first sound signal and the first decibel value when the target speaker is running, determine whether there is user voice in the sound signal, if there is, discard the first decibel value, if there is no, retain the first decibel value, until the first decibel value is retained; Step 120: After a preset time, acquire the second sound signal and the second decibel value of the target speaker during operation, determine whether there is user voice in the sound signal, discard the second decibel value if it exists, retain the second decibel value if it does not exist, until the second decibel value is retained. Step 130: Calculate the difference between the second decibel value and the first decibel value to obtain the decibel change value; Step 140: Determine the change in ambient noise based on the decibel change value. If the ambient noise changes from increasing to decreasing, trigger the voice recognition function, obtain the user's feedback voice, and adjust the speaker volume accordingly.

[0017] In this embodiment 1, by acquiring the first and second decibel values ​​of the noise-free sampling intervals respectively, and using the difference between the second and first decibel values ​​to approximate the change in environmental noise, the voice recognition function is triggered when the environmental noise changes, and voice interaction is performed with the user to adaptively adjust the speaker volume. On the one hand, the volume can be adjusted in a timely manner to suit the user's listening experience when the noise fluctuates, and on the other hand, the interaction with the user increases the user experience.

[0018] In a preferred embodiment of the present invention, specifically, determining whether user voice exists in the sound signal includes... The audio signal is passed through an FIR low-pass filter to remove high-frequency aliasing noise. Then, gain calibration is performed to convert the electrical signal amplitude into a sound pressure-related uniform scale. After that, the VAD algorithm is used to determine whether the calculated frame energy and spectral flatness meet the corresponding judgment rules. If they meet the rules, it is determined that there is user speech in the audio signal. If they do not meet the rules, it is determined that there is no user speech in the audio signal.

[0019] In this preferred embodiment, the classic VAD algorithm is used to determine whether there is human voice interference in the sound signal for obtaining the decibel value, which can accurately find the situation where human voice interference exists.

[0020] In a preferred embodiment of the present invention, specifically, determining the change in environmental noise based on the decibel change includes: A reasonable range for decibel changes is preset. If the decibel change exceeds the maximum value of the reasonable range, the environmental noise change is determined to be increasing. If the decibel change exceeds the minimum value of the reasonable range, the environmental noise change is determined to be decreasing.

[0021] In this preferred embodiment, it is assumed that the decibel level of the sound emitted by the speaker at the same volume is the same (although there may be errors due to differences in content, an approximate estimate can be made). Therefore, by subtracting the decibel values, the influence of the speaker in the two decibel values ​​is eliminated, and what remains is only the influence of environmental noise on the decibel value. Therefore, this is used as a representation of the change in environmental noise for subsequent judgment, and a more accurate judgment result can be obtained.

[0022] In a preferred embodiment of the present invention, specifically, the process of triggering the voice recognition function, obtaining the user's feedback voice, and adaptively adjusting the speaker volume based on this feedback includes: If the ambient noise changes and increases, the user is asked whether to adjust the speaker volume. If the user confirms the adjustment, the system uses (a, b) as the adaptive volume adjustment range to interact with the user via voice and thus adaptively increase the speaker volume. Here, a is the current volume and b is the preset maximum volume. If the ambient noise decreases, the user is asked whether to adjust the speaker volume. If the user confirms the adjustment, the system uses [c, a) as the adaptive volume adjustment range and interacts with the user via voice to adaptively reduce the speaker volume. c is the preset minimum volume value.

[0023] In a preferred embodiment of the present invention, specifically, the process of using (a, b) as the volume adaptive adjustment range to interact with the user via voice and thereby adaptively increasing the speaker volume includes the following steps: A volume value x is randomly selected from (a, b) as the initial volume adjustment, and user voice feedback is obtained. The user voice feedback is then subjected to speech recognition. If the speech recognition result is confirmation, the initial volume adjustment is used as the final volume. If the speech recognition result indicates that the volume needs to be increased, (x, b) is used as the new adaptive volume adjustment range, and the above steps are performed. If the speech recognition result indicates that the volume needs to be decreased, (a, x) is used as the new adaptive volume adjustment range, and the above steps are performed. If the speech recognition result indicates a vague intention, [xy, x) ∩ (x, x+y) is used as the new adaptive volume adjustment range, and the above steps are performed. y is a preset adjustment parameter, until the speech recognition result is confirmation.

[0024] In this preferred embodiment, the binary search method is used as the core idea for the search. By combining the above method with the user's voice feedback, the appropriate speaker volume that matches the user's intention can be quickly found. The whole process involves little computation and can meet the needs of application scenarios where high precision is not required.

[0025] In a preferred embodiment of the present invention, specifically, the process of using [c, a) as the volume adaptive adjustment range to interact with the user via voice and thereby adaptively reducing the speaker volume includes, A volume value p is randomly selected from [c, a) as the initial volume adjustment, and user voice feedback is obtained. The user voice feedback is then subjected to speech recognition. If the speech recognition result is confirmation, the initial volume adjustment is used as the final volume. If the speech recognition result indicates that the volume needs to be increased, (p, a) is used as the new adaptive volume adjustment range, and the above steps are performed. If the speech recognition result indicates that the volume needs to be decreased, [c, p) is used as the new adaptive volume adjustment range, and the above steps are performed. If the speech recognition result indicates a vague intention, [pq, p) ∩ (p, p+q] is used as the new adaptive volume adjustment range, and the above steps are performed, where q is a preset adjustment parameter, until the speech recognition result is confirmation.

[0026] In a preferred embodiment of the present invention, the process of obtaining speech recognition results includes: A first dictionary, a second dictionary, and a third dictionary are pre-built. When the user's speech content obtained from speech recognition matches the first dictionary, the speech recognition result is determined to be "confirmed". When the user's speech content obtained from speech recognition matches the second dictionary, the speech recognition result is determined to be "needs to increase". When the user's speech content obtained from speech recognition matches the third dictionary, the speech recognition result is determined to be "needs to decrease". If the user's speech content obtained from speech recognition does not match any word in the above three dictionaries, the speech recognition result is determined to be "fuzzy intention".

[0027] In this preferred embodiment, the user's intention to adjust the speaker volume is quickly obtained by pre-constructing a dictionary and matching the user's voice content with the dictionary.

[0028] In a preferred embodiment of the present invention, the FIR low-pass filter used is of the 64th order and has a cutoff frequency of 22kHz.

[0029] In this preferred embodiment, the use of the above-mentioned FIR low-pass filter can effectively filter out error noise, making the subsequent analysis results more accurate.

[0030] Example 2: This invention also proposes a smart speaker volume adjustment device that combines environmental noise monitoring and voice recognition, comprising the following: The first data acquisition module is used to acquire the first sound signal and the first decibel value when the target speaker is running, determine whether there is user voice in the sound signal, discard the first decibel value if it exists, and retain the first decibel value if it does not exist, until the first decibel value is retained. The second data acquisition module is used to acquire the second sound signal and the second decibel value of the target speaker after a preset time, determine whether there is user voice in the sound signal, discard the second decibel value if it exists, and retain the second decibel value if it does not exist, until the second decibel value is retained. The difference calculation module is used to calculate the difference between the second decibel value and the first decibel value to obtain the decibel change value. The volume adjustment module is used to determine the change in ambient noise based on the decibel change value. If the ambient noise changes to increase or decrease, the voice recognition function is triggered to obtain the user's feedback voice and adjust the speaker volume adaptively accordingly.

[0031] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0032] If the integrated module is implemented as a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or system capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc.

[0033] Although the description of the invention has been quite detailed and particularly of several described embodiments, it is not intended to limit it to any of these details or embodiments or any particular embodiment, but should be considered as providing a broad possible interpretation of the claims by referring to the appended claims and taking into account the prior art, thereby effectively covering the intended scope of the invention. Furthermore, the invention has been described above with respect to embodiments foreseeable by the inventors in order to provide a useful description, and non-substantial modifications to the invention that have not yet been foreseen may still represent equivalent modifications.

[0034] The above description is merely a preferred embodiment of the present invention. The present invention is not limited to the above-described embodiments. Any embodiment that achieves the technical effects of the present invention using the same means should fall within the protection scope of the present invention. Within the protection scope of the present invention, various modifications and variations can be made to the technical solutions and / or implementation methods.

Claims

1. A method for intelligent adjustment of speaker volume combining environmental noise monitoring and speech recognition, characterized in that, Including the following: The system acquires the first sound signal and the first decibel value of the target speaker during operation, determines whether user voice exists in the sound signal, discards the first decibel value if it exists, and retains the first decibel value if it does not exist, until the first decibel value is retained. After a preset duration, the second sound signal and the second decibel value of the target speaker during operation are acquired, and it is determined whether there is user voice in the sound signal. If there is, the second decibel value is discarded; if not, the second decibel value is retained until the second decibel value is retained. The difference between the second decibel value and the first decibel value is used to obtain the decibel change value. Based on the decibel change value, the environmental noise change is determined. If the environmental noise changes to increase or decrease, the voice recognition function is triggered to obtain the user's feedback voice and adjust the speaker volume accordingly.

2. The intelligent speaker volume adjustment method combining environmental noise monitoring and speech recognition according to claim 1, characterized in that, Specifically, determining whether user voice exists in the sound signal includes, The audio signal is passed through an FIR low-pass filter to remove high-frequency aliasing noise. Then, gain calibration is performed to convert the electrical signal amplitude into a sound pressure-related uniform scale. After that, the VAD algorithm is used to determine whether the calculated frame energy and spectral flatness meet the corresponding judgment rules. If they meet the rules, it is determined that there is user speech in the audio signal. If they do not meet the rules, it is determined that there is no user speech in the audio signal.

3. The intelligent speaker volume adjustment method combining environmental noise monitoring and speech recognition according to claim 1, characterized in that, Specifically, judging the change in environmental noise based on the decibel change includes, A reasonable range for decibel changes is preset. If the decibel change exceeds the maximum value of the reasonable range, the environmental noise change is determined to be increasing. If the decibel change exceeds the minimum value of the reasonable range, the environmental noise change is determined to be decreasing.

4. The intelligent speaker volume adjustment method combining environmental noise monitoring and speech recognition according to claim 3, characterized in that, Specifically, the process of triggering the voice recognition function, obtaining the user's feedback voice, and adaptively adjusting the speaker volume based on this includes: If the ambient noise changes and increases, the user is asked whether to adjust the speaker volume. If the user confirms the adjustment, the system uses (a, b) as the adaptive volume adjustment range to interact with the user via voice and thus adaptively increase the speaker volume. Here, a is the current volume and b is the preset maximum volume. If the ambient noise decreases, the user is asked whether to adjust the speaker volume. If the user confirms the adjustment, the system uses [c, a) as the adaptive volume adjustment range and interacts with the user via voice to adaptively reduce the speaker volume. c is the preset minimum volume value.

5. The intelligent speaker volume adjustment method combining environmental noise monitoring and speech recognition according to claim 4, characterized in that, Specifically, the process of using (a, b) as the volume adaptive adjustment range to interact with the user via voice and thus adaptively increase the speaker volume includes the following steps: A volume value x is randomly selected from (a, b) as the initial volume adjustment, and user voice feedback is obtained. The user voice feedback is then subjected to speech recognition. If the speech recognition result is confirmation, the initial volume adjustment is used as the final volume. If the speech recognition result indicates that the volume needs to be increased, (x, b) is used as the new adaptive volume adjustment range, and the above steps are performed. If the speech recognition result indicates that the volume needs to be decreased, (a, x) is used as the new adaptive volume adjustment range, and the above steps are performed. If the speech recognition result indicates a vague intention, [xy, x) ∩ (x, x+y) is used as the new adaptive volume adjustment range, and the above steps are performed. y is a preset adjustment parameter, until the speech recognition result is confirmation.

6. The intelligent speaker volume adjustment method combining environmental noise monitoring and speech recognition according to claim 5, characterized in that, Specifically, the process of using [c, a) as the volume adaptive adjustment range to interact with the user via voice and thus adaptively reduce the speaker volume includes: A volume value p is randomly selected from [c, a) as the initial volume adjustment, and user voice feedback is obtained. The user voice feedback is then subjected to speech recognition. If the speech recognition result is confirmation, the initial volume adjustment is used as the final volume. If the speech recognition result indicates that the volume needs to be increased, (p, a) is used as the new adaptive volume adjustment range, and the above steps are performed. If the speech recognition result indicates that the volume needs to be decreased, [c, p) is used as the new adaptive volume adjustment range, and the above steps are performed. If the speech recognition result indicates a vague intention, [pq, p) ∩ (p, p+q] is used as the new adaptive volume adjustment range, and the above steps are performed, where q is a preset adjustment parameter, until the speech recognition result is confirmation.

7. The intelligent speaker volume adjustment method combining environmental noise monitoring and speech recognition according to claim 6, characterized in that, Specifically, the process of obtaining speech recognition results includes: A first dictionary, a second dictionary, and a third dictionary are pre-built. When the user's speech content obtained from speech recognition matches the first dictionary, the speech recognition result is determined to be "confirmed". When the user's speech content obtained from speech recognition matches the second dictionary, the speech recognition result is determined to be "needs to increase". When the user's speech content obtained from speech recognition matches the third dictionary, the speech recognition result is determined to be "needs to decrease". If the user's speech content obtained from speech recognition does not match any word in the above three dictionaries, the speech recognition result is determined to be "fuzzy intention".

8. The intelligent speaker volume adjustment method combining environmental noise monitoring and speech recognition according to claim 2, characterized in that, Specifically, the FIR low-pass filter used is a 64th order filter with a cutoff frequency of 22kHz.

9. A smart speaker volume adjustment device combining environmental noise monitoring and voice recognition, characterized in that, Including the following: The first data acquisition module is used to acquire the first sound signal and the first decibel value when the target speaker is running, determine whether there is user voice in the sound signal, discard the first decibel value if it exists, and retain the first decibel value if it does not exist, until the first decibel value is retained. The second data acquisition module is used to acquire the second sound signal and the second decibel value of the target speaker after a preset time, determine whether there is user voice in the sound signal, discard the second decibel value if it exists, and retain the second decibel value if it does not exist, until the second decibel value is retained. The difference calculation module is used to calculate the difference between the second decibel value and the first decibel value to obtain the decibel change value. The volume adjustment module is used to determine the change in ambient noise based on the decibel change value. If the ambient noise changes to increase or decrease, the voice recognition function is triggered to obtain the user's feedback voice and adjust the speaker volume adaptively accordingly.

Images