Voice noise reduction method and device, storage medium and electronic equipment

Abstract

The embodiment of the invention provides a voice noise reduction method and device, a storage medium and electronic equipment. The method comprises the following steps: acquiring a voice signal to be subjected to noise reduction; determining a target frequency threshold, wherein the target frequency threshold and the sound frequency of the user meet a preset corresponding relation; and filtering a noise signal from the to-be-denoised voice signal based on the target frequency threshold to obtain a target voice signal, the frequency value of the noise signal being higher than the target frequency threshold. According to the technical scheme provided by the embodiment of the invention, the accuracy of noise signal filtering can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of noise reduction, and in particular to a speech noise reduction method, device, storage medium and electronic device. Background Art

[0002] The frequency of human voice is generally 100-1000Hz, while noise is generally high-frequency noise, with a frequency value much higher than 1000Hz. Therefore, in the noise reduction algorithm based on frequency filtering noise in the prior art, a fixed frequency value is usually selected to filter high-frequency noise signals, such as 10000Hz. It can be understood that this method can only filter high-frequency noise, but cannot filter noise below the fixed frequency value. Based on this, how to improve the accuracy of filtering noise signals is a technical problem that needs to be solved urgently. Summary of the invention

[0003] The embodiments of the present application provide a speech noise reduction method, device, storage medium and electronic device. The technical solution provided in the present application can improve the accuracy of filtering noise signals.

[0004] Other features and advantages of the present application will become apparent from the following detailed description, or may be learned in part by the practice of the present application.

[0005] According to a first aspect of an embodiment of the present application, a speech noise reduction method is provided, the method comprising: obtaining a speech signal to be denoised; determining a target frequency threshold, the target frequency threshold and the user's voice frequency satisfying a preset correspondence; filtering a noise signal from the speech signal to be denoised based on the target frequency threshold to obtain a target speech signal, the frequency value of the noise signal being higher than the target frequency threshold.

[0006] In some embodiments of the present application, based on the aforementioned scheme, determining the target frequency threshold includes: obtaining a frequency change curve, which records the changing pattern of the user's voice frequency within a cycle; based on the acquisition time of the voice signal to be denoised, determining the target frequency threshold through the frequency change curve.

[0007] In some embodiments of the present application, based on the aforementioned scheme, the method further includes: for each user, collecting the sound frequency of each user at different times within at least one cycle; based on the sound frequency of each user at different times within at least one cycle, fitting a frequency change curve matching each user.

[0008] In some embodiments of the present application, based on the aforementioned scheme, the obtaining of the frequency change curve includes: determining the user identity according to the speech signal to be denoised; and obtaining a frequency change curve matching the user identity from predetermined frequency change curves.

[0009] In some embodiments of the present application, based on the aforementioned scheme, the method further includes: for each user type, collecting the sound frequencies of each user that meets each user type at different times within at least one cycle; based on the sound frequencies of each user that meets each user type at different times within at least one cycle, fitting a frequency change curve that matches each user type.

[0010] In some embodiments of the present application, based on the aforementioned scheme, obtaining the frequency change curve includes: determining the target user type according to the speech signal to be denoised; and obtaining a frequency change curve matching the target user type from predetermined frequency change curves.

[0011] In some embodiments of the present application, based on the aforementioned scheme, the method further includes: for each environment, collecting the sound frequency of each user in each environment at different times within at least one cycle, and determining the average sound frequency of each user; taking the user with the highest average sound frequency as the target user, and fitting a frequency change curve matching each environment based on the sound frequency of the target user at different times within at least one cycle.

[0012] In some embodiments of the present application, based on the aforementioned scheme, the target frequency threshold is determined based on the acquisition time of the speech signal to be denoised through the frequency change curve, including: based on the acquisition time of the speech signal to be denoised, calibrating the target frequency value in the frequency change curve; adjusting the target frequency value according to a preset adjustment ratio to obtain the target frequency threshold, and the preset adjustment ratio is greater than or equal to 1.

[0013] According to a second aspect of an embodiment of the present application, a speech noise reduction device is provided, the device comprising: an acquisition unit, used to acquire a speech signal to be denoised; a determination unit, used to determine a target frequency threshold, the target frequency threshold and the user's voice frequency satisfy a preset correspondence; a filtering unit, used to filter a noise signal from the speech signal to be denoised based on the target frequency threshold to obtain a target speech signal, the frequency value of the noise signal being higher than the target frequency threshold.

[0014] According to a third aspect of an embodiment of the present application, a computer-readable storage medium is provided, characterized in that at least one program code is stored in the computer-readable storage medium, and the at least one program code is loaded and executed by a processor to implement the operations performed by the method described in any one of the first aspects above.

[0015] According to a fourth aspect of an embodiment of the present application, an electronic device is provided, comprising one or more processors and one or more memories, wherein at least one program code is stored in the one or more memories, and the at least one program code is loaded and executed by the one or more processors to implement the operations performed by the method described in any one of the first aspects above.

[0016] The technical solution of the present application first obtains the voice signal to be de-noised; then determines the target frequency threshold that satisfies a preset correspondence with the user's voice frequency; and finally filters the noise signal from the voice signal to be de-noised based on the target frequency threshold to obtain the target voice signal, wherein the noise signal frequency value is higher than the target frequency threshold. It can be seen that the technical solution of the present application uses the target frequency threshold that satisfies a preset correspondence with the user's voice frequency to filter the noise signal in the acquired voice signal to be de-noised. Compared with the prior art that uses the same frequency threshold to filter the noise signal for any user, the technical solution of the present application is based on the target frequency threshold customized for the user's voice, so that different users can use different target frequency thresholds to filter the noise signal, thereby retaining the user's real voice signal to the greatest extent in the process of filtering the noise, filtering the noise signal in the voice signal to be de-noised to the greatest extent, thereby improving the accuracy of filtering the noise signal, and improving the user experience.

[0017] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The drawings herein are incorporated into the specification and constitute a part of the specification, showing embodiments consistent with the present application, and together with the specification, are used to explain the principles of the present application. Obviously, the drawings described below are only some embodiments of the present application, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0019] Figure 1 A schematic diagram of a flow chart of a method for reducing speech noise according to an embodiment of the present application is shown;

[0020] Figure 2 A detailed flow chart of determining a target frequency threshold according to an embodiment of the present application is shown;

[0021] Figure 3 A schematic diagram showing a frequency variation curve according to an embodiment of the present application is shown;

[0022] Figure 4 A block diagram of a speech noise reduction device according to an embodiment of the present application is shown;

[0023] Figure 5 A schematic structural diagram of an electronic device according to an embodiment of the present application is shown. DETAILED DESCRIPTION

[0024] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this application will be more comprehensive and complete and fully convey the concept of the example embodiments to those skilled in the art.

[0025] In addition, described feature, structure or characteristic can be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present application. However, those skilled in the art will appreciate that the technical scheme of the present application can be put into practice without one or more of the specific details, or other methods, components, devices, steps, etc. can be adopted. In other cases, known methods, devices, realizations or operations are not shown or described in detail to avoid blurring the various aspects of the application.

[0026] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities may be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0027] The flowcharts shown in the accompanying drawings are only exemplary and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps can be decomposed, and some operations / steps can be combined or partially combined, so the actual execution order may change according to actual conditions.

[0028] It should be noted that the "multiple" mentioned in this article refers to two or more. "And / or" describes the association relationship of the associated objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the associated objects before and after are in an "or" relationship.

[0029] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the objects used in this way can be interchanged where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those shown or described.

[0030] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is only a part of the embodiment of the present invention, not all the embodiments. Based on the embodiment of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0031] Some embodiments of the present application will be described in detail below in conjunction with the accompanying drawings. In the absence of conflict, the following embodiments and features in the embodiments may be combined with each other.

[0032] See also Figure 1 , shows a flow chart of a speech noise reduction method according to an embodiment of the present application, which specifically includes S110 to S130:

[0033] S110, obtaining a speech signal to be denoised.

[0034] It should be noted that the speech noise reduction method of the present application can be executed in smart devices that need to perform speech noise reduction processing, such as smart speakers, televisions, refrigerators, air conditioners, etc. Exemplarily, the user issues a voice command to the smart device, and the smart device collects a voice signal containing the voice command issued by the user through a microphone, and then filters the noise signal in the collected voice signal through the speech noise reduction method, retaining only the audio signal issued by the user, and then converting the user's audio signal into a corresponding text or command to instruct the smart device to perform a corresponding action.

[0035] Therefore, in step S110, the smart device that needs to perform voice noise reduction processing may obtain a voice signal to be noise-reduced. The voice signal to be noise-reduced may be a voice signal currently emitted by the user, or a voice signal recorded in advance by the user, which is not limited in this application.

[0036] Continue to see Figure 1 , S120, determining a target frequency threshold, wherein the target frequency threshold and the user's voice frequency satisfy a preset corresponding relationship.

[0037] In step S120, the specific implementation of determining the target frequency threshold includes at least the following two implementations.

[0038] The first embodiment includes the following steps S10 to S12:

[0039] S10: determining a user identity according to the speech signal to be denoised.

[0040] S11, determining an upper limit sound frequency value that matches the user identity from a plurality of pre-stored upper limit sound frequency values ​​as a target upper limit sound frequency value.

[0041] S12, adjusting the target upper limit sound frequency value according to a preset adjustment ratio to obtain the target frequency threshold, wherein the preset adjustment ratio is greater than or equal to 1. It can be understood that the purpose of this step is to prevent the voice signal sent by the user from being filtered.

[0042] The plurality of upper limit sound frequency values ​​stored in step S11 may be obtained through the following steps S1 to S2 .

[0043] S1, for each user, collecting the sound frequency of each user at different times.

[0044] S2, determining an upper limit sound frequency corresponding to each user based on the sound frequency of each user at different times.

[0045] It can be understood that the upper limit sound frequency is the upper limit value that the user's sound frequency can reach. For example, if it is determined that the upper limit sound frequency value of user A is 1000 Hz, the preset adjustment ratio is set to 1.1, so the target frequency threshold can be determined to be 1100 Hz.

[0046] The second embodiment can be as follows Figure 2 Follow the steps shown.

[0047] See also Figure 2 , shows a detailed flow chart of determining a target frequency threshold according to an embodiment of the present application, which specifically includes the following steps S121 to S122:

[0048] S121, obtaining a frequency change curve, where the frequency change curve records a change pattern of the user's voice frequency within a cycle.

[0049] It should be noted that the one cycle can be 1 day, 1 month, 1 year, etc., and there is no specific limitation.

[0050] The specific implementation methods of acquiring the frequency variation curve in step S121 include at least the following three implementation methods.

[0051] The first implementation of step S121:

[0052] In the first embodiment, the frequency variation curve can be obtained according to the following steps S100A to S200A.

[0053] S100A, for each user, collecting the sound frequency of each user at different times within at least one cycle.

[0054] The method of collecting the user's voice frequency at different times can be that the user obtains the user's voice frequency by recording; or the smart device collects the user's voice frequency within a set time before executing the voice noise reduction method of the present application, such as collecting the user's voice frequency within two days or three days before executing the present method. Specifically, in this embodiment, there is no limitation on how to collect the user's voice frequency at different times.

[0055] In some implementations, each user may be a user that needs to be contacted in the area served by the smart device. For example, assuming that air conditioner A exists in family A, the sound frequencies of each member of family A may be collected.

[0056] S200A: Fit a frequency change curve matching each user based on the sound frequency of each user at different times within at least one cycle.

[0057] It can be understood that a frequency variation curve matching each user can be obtained by fitting based on the collected sound frequency data of each user.

[0058] In order to make those skilled in the art better understand this embodiment, Figure 3 The frequency variation curve is described in detail.

[0059] See also Figure 3 , shows a schematic diagram of a frequency variation curve according to an embodiment of the present application.

[0060] Figure 3 The horizontal axis represents time, and the vertical axis represents sound frequency. According to collecting the sound frequency of user A at different times in at least one day, at least the following Figure 3 The scattered sound frequency data shown in the figure are fitted to obtain the following Figure 3 The frequency change curve shown in FIG. 1 represents the change pattern of the sound frequency of user A in one day. It can be understood that in this curve, each time point corresponds to a sound frequency value.

[0061] Based on the above steps S100A to S200A, a frequency change curve matching each user is obtained. In the first implementation, a specific implementation method of obtaining the frequency change curve can be performed according to the following steps S1211A to S1212A:

[0062] S1211A: Determine the user identity according to the voice signal to be denoised.

[0063] S1212A: Obtain a frequency change curve matching the user identity from predetermined frequency change curves.

[0064] Exemplarily, assuming that air conditioner A is installed in home A where user A is located, air conditioner A executes the voice noise reduction method of the present application, and obtains a frequency change curve matching each member in home A through the above S100A to S200A. After air conditioner A receives the voice signal to be noise reduced sent by user A, it can determine the frequency change curve matching user A from the pre-stored frequency change curves.

[0065] In the first implementation of step S121, a matching frequency change curve is customized for each user according to the voice frequency characteristics, so that in the subsequent noise filtering process, noise filtering can be fully matched with the user's voice frequency characteristics, so that the user's voice signal is basically retained after filtering, thereby improving the accuracy of filtering noise signals.

[0066] The second implementation of step S121:

[0067] In the second embodiment, the frequency variation curve can be obtained according to the following steps S100B to S200B.

[0068] S100B, for each user type, collecting the sound frequencies of each user that matches the user type at different times within at least one cycle.

[0069] It should be noted that user types can be divided according to gender, such as male as one user type and female as another user type; user types can also be divided according to age, such as the elderly as one user type, the middle-aged as one user type, the teenagers as one user type, and the children as one user type. Specifically, user types can also be divided according to other methods, which are not limited in this application.

[0070] S200B: fitting a frequency change curve matching each user type based on the sound frequencies of each user matching each user type at different times within at least one cycle.

[0071] In some implementations, for each user type, the frequency change curve of each user under the user type may be fitted first, and then the frequency change curve fitted for each user may be further curve fitted to finally obtain a frequency change curve matching the user type.

[0072] For example, assuming that user types are divided according to gender, assuming that the sound frequencies of male A, male B, female A, and female B at different times of at least one day are collected respectively, so that the frequency change curves of male A, male B, female A, and female B are fitted respectively based on the collected sound frequency data. Further, curve fitting can be performed based on the frequency change curves of male A and male B to obtain the frequency change curve of male user type; curve fitting can be performed based on the frequency change curves of female A and female B to obtain the frequency change curve of female user type.

[0073] In the process of curve fitting the frequency change curve, the maximum sound frequency value corresponding to the same time can be used as the sound frequency value of the frequency curve finally generated. For example, the sound frequency value corresponding to the frequency change curve of male A at 12:00 is 800Hz, and the sound frequency value corresponding to the frequency change curve of male B at 12:00 is 700Hz, then 800Hz can be selected as the sound frequency value corresponding to the frequency change curve at 12:00 that matches the male user type.

[0074] Based on the above steps S100B to S200B, a frequency change curve matching each user type is obtained. In the second implementation, a specific implementation method of obtaining the frequency change curve can be performed according to the following steps S1211B to S1212B:

[0075] S1211B: Determine a target user type according to the speech signal to be denoised.

[0076] S1212B: Acquire a frequency change curve matching the target user type from predetermined frequency change curves.

[0077] Exemplarily, if it is determined that the target user type corresponding to the speech signal to be denoised is male, then the frequency change curve corresponding to male can be determined from the frequency change curves pre-obtained based on steps S100B to S200B.

[0078] A third implementation of step S121:

[0079] In the third embodiment, the frequency variation curve can be obtained according to the following steps S100C to S200C.

[0080] S100C, for each environment, collecting the sound frequencies of each user in each environment at different times within at least one cycle, and determining an average sound frequency of each user.

[0081] The environment is related to the area served by the smart device. For example, if air conditioner A serves in home A, then home A is an environment. If air conditioner A serves in office area A, then office area A is an environment.

[0082] For example, if there are user A and user B in family A, and there are user C and user D in family B, then based on step S100C, the average sound frequency values ​​of user A, user B, user C, and user D can be obtained respectively.

[0083] S200C: taking the user with the highest average sound frequency as the target user, and fitting a frequency change curve matching each environment based on the sound frequency of the target user at different times in at least one cycle.

[0084] For example, if the average sound frequency of user A is determined to be 780 Hz, the average sound frequency of user B is 860 Hz, the average sound frequency of user C is 800 Hz, and the average sound frequency of user D is 900 Hz, then for family A, the collected sound frequency data of user B can be used for fitting to obtain a frequency change curve that matches family A. For family B, the collected sound frequency data of user D can be used for fitting to obtain a frequency change curve that matches family B.

[0085] Based on the above steps S100C to S200C, a frequency change curve matching each environment is obtained. In the third embodiment, a specific implementation method of obtaining the frequency change curve may be to automatically match a frequency change curve adapted to the environment by detecting the environment in which the smart device is located. For example, if the smart device is installed in home A, a frequency change curve matching home A may be obtained.

[0086] Based on the above description, the frequency change curve of the above step S121 can be obtained based on the user identity, the user type, or the environment. It can be seen that the speech noise reduction method of the present application is based on user big data adaptive adjustment, thereby improving the accuracy of filtering noise signals.

[0087] Continue to see Figure 2 , S122, based on the acquisition time of the speech signal to be denoised, determine the target frequency threshold through the frequency change curve.

[0088] In some implementations, the specific implementation of determining the target frequency threshold through the frequency change curve based on the acquisition time of the speech signal to be denoised may be performed according to the following S1221 to S1222:

[0089] S1221: Calibrate a target frequency value in the frequency change curve based on the acquisition time of the speech signal to be denoised.

[0090] S1222: Adjust the target frequency value according to a preset adjustment ratio to obtain the target frequency threshold, where the preset adjustment ratio is greater than or equal to 1.

[0091] Among them, the preset adjustment ratio can be set to 1, 1.1, 1.2, etc., and the specific details are not limited in this application.

[0092] Exemplarily, the frequency change curve of user A is obtained in the first implementation of step S121, such as Figure 3 Assume that the acquisition time of the speech signal to be de-noised is 9:00, then Figure 3 The target frequency value is determined to be 750 Hz in the curve shown. If the selected preset adjustment ratio is 1.1, the target frequency threshold value can be determined to be 825 Hz.

[0093] In this embodiment, by setting a preset adjustment ratio, it is possible to prevent the user's real voice frequency from being filtered, thereby improving the accuracy of filtering noise.

[0094] Continue to see Figure 1 , S130, filtering the noise signal from the speech signal to be denoised based on the target frequency threshold to obtain a target speech signal, wherein the frequency value of the noise signal is higher than the target frequency threshold.

[0095] It can be understood that the target frequency threshold can be obtained through the above steps S110 to S120, so the frequency signal above the target frequency threshold in the speech signal to be denoised is filtered as noise, so that the final target speech signal is basically the user's real speech signal, thereby improving the accuracy of subsequent speech recognition.

[0096] In the technical solutions provided in some embodiments of the present application, first, a speech signal to be denoised is obtained; then a target frequency threshold that satisfies a preset correspondence with the user's voice frequency is determined; finally, based on the target frequency threshold, a noise signal is filtered from the speech signal to be denoised to obtain a target speech signal, and the frequency value of the noise signal is higher than the target frequency threshold. It can be seen that the technical solution of the present application uses a target frequency threshold that satisfies a preset correspondence with the user's voice frequency to filter the noise signal in the acquired speech signal to be denoised. Compared with the prior art that uses the same frequency threshold to filter noise signals for any user, the technical solution of the present application is based on a target frequency threshold customized for the user's voice, so that different users can use different target frequency thresholds to filter noise signals, thereby retaining the user's real voice signal to the greatest extent in the process of filtering noise, filtering the noise signal in the speech signal to be denoised to the greatest extent, thereby improving the accuracy of filtering noise signals, and improving user experience.

[0097] Based on the same inventive concept, an embodiment of the present invention provides a speech noise reduction device, which can be used to execute the speech noise reduction method in the above embodiment of the present application. For details not disclosed in the embodiment of the present application, please refer to the embodiment of the speech noise reduction method in the above embodiment of the present application.

[0098] See also Figure 4 , shows a block diagram of a speech noise reduction device according to an embodiment of the present application.

[0099] like Figure 4 As shown, a speech noise reduction device 400 according to an embodiment of the present application includes: an acquisition unit 401, a determination unit 402, and a filtering unit 403.

[0100] Among them, the acquisition unit 401 is used to acquire the speech signal to be denoised; the determination unit 402 is used to determine the target frequency threshold, and the target frequency threshold and the user's voice frequency satisfy a preset correspondence; the filtering unit 403 is used to filter the noise signal from the speech signal to be denoised based on the target frequency threshold to obtain the target speech signal, and the frequency value of the noise signal is higher than the target frequency threshold.

[0101] In some embodiments of the present application, based on the aforementioned scheme, the determination unit 402 is also used to: obtain a frequency change curve, which records the change pattern of the user's voice frequency within a cycle; based on the acquisition time of the voice signal to be denoised, determine the target frequency threshold through the frequency change curve.

[0102] In some embodiments of the present application, based on the aforementioned scheme, the determination unit 402 is also used to: for each user, collect the sound frequency of each user at different times within at least one cycle; based on the sound frequency of each user at different times within at least one cycle, fit a frequency change curve matching each user.

[0103] In some embodiments of the present application, based on the above scheme, the determination unit 402 is further used to: determine the user identity according to the speech signal to be denoised; and obtain a frequency change curve matching the user identity from predetermined frequency change curves.

[0104] In some embodiments of the present application, based on the aforementioned scheme, the determination unit 402 is further used to: for each user type, collect the sound frequencies of each user that meets each user type at different times within at least one cycle; based on the sound frequencies of each user that meets each user type at different times within at least one cycle, fit a frequency change curve that matches each user type.

[0105] In some embodiments of the present application, based on the aforementioned scheme, the determination unit 402 is further used to: determine the target user type according to the speech signal to be denoised; and obtain a frequency change curve matching the target user type from each predetermined frequency change curve.

[0106] In some embodiments of the present application, based on the aforementioned scheme, the determination unit 402 is further used to: for each environment, collect the sound frequency of each user in each environment at different times within at least one cycle, and determine the average sound frequency of each user; take the user with the highest average sound frequency as the target user, and based on the sound frequency of the target user at different times within at least one cycle, fit a frequency change curve matching each environment.

[0107] In some embodiments of the present application, based on the aforementioned scheme, the determination unit 402 is also used to: calibrate the target frequency value in the frequency change curve based on the acquisition time of the speech signal to be denoised; adjust the target frequency value according to a preset adjustment ratio to obtain the target frequency threshold, and the preset adjustment ratio is greater than or equal to 1.

[0108] Based on the same inventive concept, an embodiment of the present application provides a computer-readable storage medium, in which at least one computer program instruction is stored. The at least one computer program instruction is loaded and executed by a processor to implement the operations performed by the method described above.

[0109] Based on the same inventive concept, an embodiment of the present application also provides an electronic device.

[0110] See also Figure 5 , shows a schematic diagram of the structure of an electronic device according to an embodiment of the present application, wherein the electronic device includes one or more memories 504, one or more processors 502, and at least one computer program (computer program instruction) stored in the memory 504 and executable on the processor 502, and the processor 502 implements the method described above when executing the computer program.

[0111] Among them, Figure 5 In the embodiment of the present invention, a bus architecture (represented by bus 500) is shown, which may include any number of interconnected buses and bridges, and bus 500 links various circuits including one or more processors represented by processor 502 and memory represented by memory 504. Bus 500 may also link various other circuits such as peripherals, voltage regulators, and power management circuits, which are well known in the art and are therefore not further described herein. Bus interface 505 provides an interface between bus 500 and receiver 501 and transmitter 503. Receiver 501 and transmitter 503 may be the same element, namely a transceiver, providing a unit for communicating with various other devices over a transmission medium. Processor 502 is responsible for managing bus 500 and general processing, while memory 504 may be used to store data used by processor 502 when performing operations.

[0112] The functions described herein may be implemented in hardware, software executed by a processor, firmware, or any combination thereof. If implemented in software executed by a processor, the functions may be stored as one or more instructions or codes on a computer-readable medium or transmitted via a computer-readable medium. Other examples and implementations are within the scope and spirit of the present application and the appended claims. For example, due to the nature of software, the functions described above may be implemented using software executed by a processor, hardware, firmware, hard wiring, or a combination of any of these. In addition, each functional unit may be integrated into a processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit.

[0113] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0114] The units described as separate components may or may not be physically separated, and the components of the control device may or may not be physical units, that is, they may be located in one place or distributed in multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0115] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly or all or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or optical disk, etc., which can store computer program instructions.

[0116] The above description is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent substitution, improvement, etc. made within the spirit and principle of the present application shall be included in the scope of the claims of the present application.

Claims

1. A speech noise reduction method, characterized in that: The method comprises: Obtaining a speech signal to be de-noised; Determine a target frequency threshold, where the target frequency threshold satisfies a preset corresponding relationship with the user's voice frequency; A noise signal is filtered from the speech signal to be de-noised based on the target frequency threshold to obtain a target speech signal, wherein the frequency value of the noise signal is higher than the target frequency threshold.

2. The method according to claim 1, characterized in that The step of determining the target frequency threshold comprises: Obtaining a frequency change curve, where the frequency change curve records the change pattern of the user's voice frequency within a cycle; Based on the acquisition time of the speech signal to be denoised, a target frequency threshold is determined through the frequency change curve.

3. The method according to claim 2, characterized in that The method further comprises: For each user, collecting the sound frequency of each user at different times in at least one cycle; Based on the sound frequencies of each user at different times within at least one cycle, a frequency change curve matching each user is fitted.

4. The method according to claim 3, characterized in that The obtaining of the frequency change curve comprises: Determining the user identity according to the speech signal to be de-noised; A frequency change curve matching the user identity is obtained from each predetermined frequency change curve.

5. The method according to claim 2, characterized in that: The method further comprises: For each user type, collecting the sound frequencies of each user that matches each user type at different times within at least one cycle; Based on the sound frequencies of the users matching each user type at different times within at least one cycle, a frequency change curve matching each user type is fitted.

6. The method according to claim 5, characterized in that The obtaining of the frequency change curve comprises: Determining a target user type according to the speech signal to be de-noised; A frequency change curve matching the target user type is obtained from each predetermined frequency change curve.

7. The method according to claim 2, characterized in that The method further comprises: For each environment, collecting the sound frequencies of each user in each environment at different times within at least one cycle, and determining the average sound frequency of each user; The user with the highest average sound frequency is taken as the target user, and a frequency change curve matching each environment is fitted based on the sound frequency of the target user at different times in at least one cycle.

8. The method according to claim 2, characterized in that: The step of determining the target frequency threshold value based on the acquisition time of the speech signal to be denoised by the frequency change curve includes: Based on the acquisition time of the speech signal to be denoised, calibrating a target frequency value in the frequency change curve; The target frequency value is adjusted according to a preset adjustment ratio to obtain the target frequency threshold, and the preset adjustment ratio is greater than or equal to 1.

9. A speech noise reduction device, characterized in that: The device comprises: An acquisition unit, used for acquiring a speech signal to be denoised; A determination unit, configured to determine a target frequency threshold, wherein the target frequency threshold and the user's voice frequency satisfy a preset corresponding relationship; The filtering unit is used to filter the noise signal from the speech signal to be denoised based on the target frequency threshold to obtain a target speech signal, wherein the frequency value of the noise signal is higher than the target frequency threshold.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores at least one program code, and the at least one program code is loaded and executed by a processor to implement the operations performed by the method according to any one of claims 1 to 8.

11. An electronic device, characterized in that: The method comprises one or more processors and one or more memories, wherein at least one program code is stored in the one or more memories, and the at least one program code is loaded and executed by the one or more processors to implement the method according to any one of claims 1 to 8.

Images