WO2015085532A1 - Réduction de bruit de signal - Google Patents

Réduction de bruit de signal Download PDF

Info

Publication number
WO2015085532A1
WO2015085532A1 PCT/CN2013/089189 CN2013089189W WO2015085532A1 WO 2015085532 A1 WO2015085532 A1 WO 2015085532A1 CN 2013089189 W CN2013089189 W CN 2013089189W WO 2015085532 A1 WO2015085532 A1 WO 2015085532A1
Authority
WO
WIPO (PCT)
Prior art keywords
segment
spectral component
magnitude
magnitude value
value
Prior art date
Application number
PCT/CN2013/089189
Other languages
English (en)
Inventor
Sheng Wu
Fuhuei Lin
Jingming Xu
Bin Jiang
Original Assignee
Spreadtrum Communications (Shanghai) Co., Ltd.
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Spreadtrum Communications (Shanghai) Co., Ltd. filed Critical Spreadtrum Communications (Shanghai) Co., Ltd.
Priority to PCT/CN2013/089189 priority Critical patent/WO2015085532A1/fr
Publication of WO2015085532A1 publication Critical patent/WO2015085532A1/fr

Links

Classifications

    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
    • G10L21/00—Processing of the speech or voice signal to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
    • G10L21/02—Speech enhancement, e.g. noise reduction or echo cancellation
    • G10L21/0208—Noise filtering
    • G10L21/0216—Noise filtering characterised by the method used for estimating noise
    • G10L21/0232—Processing in the frequency domain

Definitions

  • a method for implementing signal noise reduction may include receiving an input digital signal that represents an audio signal.
  • the method may further include parsing the input digital signal into a series of adjacent segments.
  • the method may further include converting each of the adjacent segments to a frequency domain representation.
  • the method may further include comparing a particular spectral component of a particular segment, of the series of adjacent segments, to a related spectral component of a first segment adjacent the particular segment, and to a related spectral component of a second segment adjacent the particular segment.
  • the method may further include modifying the particular spectral component upon a magnitude value of the particular spectral component at least one of: exceeding a particular scaled magnitude value of the related spectral component of the first segment; and exceeding a particular scaled magnitude value of the related spectral component of the second segment.
  • a method for detecting signal noise may include receiving an input digital signal representing a segment of audio signals.
  • the method may further include converting the input digital signal to a frequency domain representation, the frequency domain representation comprising a series of frames m each comprising n spectral magnitude values identified by a particular index k.
  • the method may further include detecting a signal anomaly by determining whether a particular spectral magnitude value at a particular index of a particular frame satisfies at least one of a set conditions comprising a predetermined peaking condition, a predetermined step-up condition, and a predetermined step-down condition.
  • the method may further include modifying the particular spectral magnitude value at the particular index of the particular frame upon the particular spectral magnitude value satisfying at least one of the set of conditions.
  • the method may further include converting the frequency domain representation to an output digital signal.
  • a computing system may include a processing unit, and a system memory connected to the processing unit.
  • the system memory may include instructions that, when executed by the processing unit, cause the processing unit to instantiate at least one module to implement signal noise reduction.
  • the at least one module may be configured to receive an input digital signal that represents an audio signal.
  • the at least one module may further be configured to partition the input digital signal into a series of adjacent segments.
  • the at least one module may further be configured to map each of the adjacent segments to a frequency domain representation.
  • the at least one module may further be configured to compare a particular spectral component of a particular segment, of the series of adjacent segments, to a related spectral component of a first segment adjacent the particular segment, and to a related spectral component of a second segment adjacent the particular segment.
  • the at least one module may further be configured to modify the particular spectral component upon a magnitude value of the particular spectral component at least one of: exceeding a particular scaled magnitude value of the related spectral component of the first segment; and exceeding a particular scaled magnitude value of the related spectral component of the second segment.
  • FIG. 1 shows an example method for implementing signal noise reduction in accordance with the present disclosure.
  • FIG. 2 shows an example segment of an example audio stream sample, and the example segment of the example audio stream sample following signal noise reduction.
  • FIG. 4 shows an example computing system or device.
  • the present disclosure is generally directed towards signal noise reduction. Although described in the context of audio signals throughout, the various aspects of the present disclosure may be applicable to any scenario in which it is desirable to implement computationally inexpensive signal processing techniques to filter undesired signal noise, as opposed to attempting to address the issue of signal noise by optimizing physical aspects of a particular system such as, for example, physical connection/connector optimization, circuit architecture optimization, and others. Optimizing physical aspects of a particular system may be at least somewhat effective on an application-by-application, or case-by-case, basis. However, it may be difficult to address signal noise reduction across a broad spectrum by investing resources to optimize physical system aspects alone. Signal processing on the other hand is extensible, or portable, in the sense that a core noise reduction algorithm, such as described below, may generally be incorporated within any system to implement signal noise reduction, aside from modifications that may be necessary due to implementation-specific requirements in software or firmware.
  • the example method 100 may be wholly or at least partially implemented by a special-purpose computing system or device, which itself may, in some embodiments, be configured based on implementation-specific requirements or specifications.
  • a computing system or device is described in further detail below in connection with FIG. 4.
  • an input audio signal x(i) which may or may not be pulse-code modulated depending on implementation-specific requirements, having an interframe interval L composed of frame length N vector may be transformed into the frequency domain according to an example windowed DFT (Discrete Fourier Transform) function of the form:
  • DFT frequency-domain transformation
  • other embodiments of frequency-domain transformation are possible such as, for example, DCT/iDCT (Discrete Cosine Transform/Inverse Discrete Cosine Transform), MDCT/iMDCT (Modified Discrete Cosine Transform/Inverse Modified Discrete Cosine Transform), and many others.
  • DFT, DCT, and/or MDCT may be mixed with a polyphase filter bank to obtain frequency-domain transformation. It is contemplated that any method for transforming a signal to/from a frequency- domain representation may be performed, and is within the scope of the present disclosure. Further, a particular method for transforming a signal to/from a frequency-domain representation may or may not be implementation-specific.
  • X A may correspond to a spectral amplitude
  • Xp may correspond to a spectral phase frequency
  • the variable N may correspond to an even, non-zero integer value.
  • /J understood to represent sampling frequency, and sampling ranges defined as [0.004 s , 0.01 1]
  • a comparison of respective spectral amplitudes at a particular index value k may be implemented to determine whether a signal anomaly such as, for example, high frequency transient noise, or so-called "clicking" noise, is present in the current frame n.
  • a signal anomaly such as, for example, high frequency transient noise, or so-called "clicking" noise
  • Such an anomaly may be injected into a particular signal by many different types of mechanisms such as, for example, connection/disconnection of particular electrical equipment, electrical component switching, and many others.
  • a signal anomaly may be determined to present in the current frame n upon satisfaction of at least one of the following conditions, presented in a mathematical form:
  • a signal anomaly may be determined to be present in the current frame n when spectral magnitude XA at index k of the current frame n is greater than 200% of the spectral magnitude XA at index k of the preceding frame n- ⁇ , and greater than 200% of the spectral magnitude XA at index k of the subsequent frame n+l (e.g., Equation 4).
  • such an implementation may be referred to as a "peaking" condition or a "predetermined peaking condition.”
  • a signal anomaly may further be determined to be present in the current frame n when spectral magnitude XA at index k of the current frame n is greater than 400% of the spectral magnitude XA at index k of the preceding frame n-l (e.g., Equation 5).
  • such an implementation may be referred to as a "step-up” condition or a "predetermined step-up condition.”
  • a signal anomaly may still further be determined to be present in the current frame n when spectral magnitude X A at index k of the current frame n is greater than 400% of the spectral magnitude XA at index k of the subsequent frame n+l (e.g., Equation 6).
  • such an implementation may be referred to as a "step-down” condition or a "predetermined step-down condition.”
  • Other embodiments are possible.
  • Equations 4-6 enable for a determination to be made as to whether a particular frequency component at index k changes too “rapidly” or “fast” on frame- by-frame basis.
  • the current frame n, and associated portion of the signal x(i) as described below in connection with FIG. 2 includes high frequency transient noise, or "clicking" noise.
  • a "threshold” or “sensitivity” that defines whether or not the particular frequency component at index k changes too “rapidly” or “fast” may be adjusted as desired via manipulation of the scalars X a , Xb, X c , and Xa. Other embodiments are possible as well.
  • any number of frames preceding or subsequent to the current frame n may be analyzed to determine whether a signal anomaly is present, or is "likely" present, in the current frame n.
  • a comparison of respective spectral amplitudes at a particular index value k may be implemented to determine whether a signal anomaly is present in the current frame n.
  • a signal anomaly may be determined to present in the current frame n upon satisfaction of at least one of the following conditions, presented in a mathematical form:
  • Equation 8 x A n+2 ⁇ x A n
  • a signal anomaly may be determined to be present in the current frame n when spectral magnitude X A at index k of the current frame n is greater than 800% of the spectral magnitude X A at index k of the preceding frame n-2, and greater than 800% of the spectral magnitude X A at index k of the subsequent frame n+2 (e.g., Equation 7).
  • a signal anomaly may further be determined to be present in the current frame n when spectral magnitude X A at index k of the current frame n is greater than 1600% of the spectral magnitude X A at index k of the preceding frame n-2 (e.g., Equation 8).
  • a signal anomaly may still further be determined to be present in the current frame n when spectral magnitude X A at index k of the current frame n is greater than 1600% of the spectral magnitude X A at index k of the subsequent frame n+2 (e.g., Equation 9).
  • Other embodiments are possible.
  • a comparison of respective spectral amplitude at a particular index value k may be implemented to determine whether a signal anomaly such as, for example, high frequency transient noise, or so-called “clicking" noise, is “likely” present or “possibly” present in the current frame n.
  • a signal anomaly such as, for example, high frequency transient noise, or so-called “clicking" noise
  • particular spectral components of the current frame n may or may not be filtered based on the comparison.
  • the spectral magnitude X A at index k of the current frame n may remain in unmodified form as X ⁇ , in accordance with an identity function of the form:
  • the spectral magnitude X A at index k of the current frame n may be modified in accordance with a particular signal conditioning function, as part of a signal noise reduction process according to the present disclosure.
  • the particular signal conditioning function may itself be a function of the number of frames preceding or subsequent to the current frame n that have been analyzed to determine whether a signal anomaly is present in the current frame n.
  • the spectral magnitude X A at index k of the current frame n may be modified as ⁇ to a magnitude corresponding to the lesser of, or minimum of, the spectral magnitude X A at index k of the preceding frame n-1, and the spectral magnitude X A at index k of the subsequent frame n+1.
  • Equation 11 corresponds to a "minimum" function, which includes arguments consistent with the number of frames preceding or subsequent to the current frame n that have been analyzed (e.g., operation 104) to determine whether a signal anomaly is present in the current frame n.
  • Equation 11 specifies the frames n-1 and n+1, such as described above in connection with Equations 4-6.
  • the spectral magnitude X A at index k of the current frame n may be modified in accordance with a signal conditioning function of the form:
  • the spectral magnitude X A at index k of the current frame n may be modified as ⁇ to a magnitude corresponding to the lesser of, or minimum of, the spectral magnitude X A at index k of the preceding frame n-2, the spectral magnitude X A at index k of the preceding frame n-1, the spectral magnitude X A at index k of the subsequent frame n+1, and the spectral magnitude X A at index k of the subsequent frame n+2.
  • Equation 12 corresponds to a "minimum" function, which includes arguments consistent with the number of frames preceding or subsequent to the current frame n that have been analyzed (e.g., operation 104) to determine whether a signal anomaly is present in the current frame n.
  • Equation 12 specifies the frames n-1, n-2, n+1, and n+2, such as described above in connection with Equations 7-9.
  • the spectral magnitude X A at index k of the current frame n may be modified in accordance with a signal conditioning function of the form: 2
  • the spectral magnitude X A at index k of the current frame n may be modified as X ⁇ to a magnitude corresponding to an arithmetic average of the spectral magnitude X A at index k of the preceding frame n-1, and the spectral magnitude X A at index k of the subsequent frame n+1.
  • Equation 13 corresponds to an "averaging" function, which includes arguments consistent with the number of frames preceding or subsequent to the current frame n that have been analyzed (e.g., operation 104) to determine whether a signal anomaly is present in the current frame n.
  • Equation 13 specifies the frames n-1 and n+1, such as described above in connection with Equations 4-6.
  • the spectral magnitude X A at index k of the current frame n may be modified in accordance with a signal conditioning function of the form: x A [k _ 2 + x A [*]._ ! + x A m n+1 + x A [*]
  • the spectral magnitude X A at index k of the current frame n may be modified as X ⁇ to a magnitude corresponding to an arithmetic average of the spectral magnitude X A at index k of the preceding frame n-2, the spectral magnitude X A at index k of the preceding frame n-1, the spectral magnitude X A at index k of the subsequent frame n+1, and the spectral magnitude X A at index k of the subsequent frame n+2.
  • Equation 14 corresponds to an "averaging" function, which includes arguments consistent with the number of frames preceding or subsequent to the current frame n that have been analyzed (e.g., operation 104) to determine whether a signal anomaly is present in the current frame n.
  • Equation 14 specifies the frames n-1, n-2, n+1, and n+2, such as described above in connection with Equations 7-9.
  • the spectral magnitude X A at index k of the current frame n may be modified in accordance with a signal conditioning function of the form: log (X A [3 ⁇ 4_ 1 ) + log (X A [ ⁇
  • the spectral magnitude XA at index k of the current frame n may be modified as ⁇ ⁇ to a magnitude corresponding to a geometric average of the spectral magnitude XA at index k of the preceding frame n-1, and the spectral magnitude XA at index k of the subsequent frame n+1.
  • Equation 15 corresponds to an "averaging" function, which includes arguments consistent with the number of frames preceding or subsequent to the current frame n that have been analyzed (e.g., operation 104) to determine whether a signal anomaly is present in the current frame n.
  • Equation 15 specifies the frames n-1 and n+1, such as described above in connection with Equations 4-6.
  • the spectral magnitude XA at index k of the current frame n may be modified in accordance with a signal conditioning function of the form: log (X A [k ⁇ _ 2 ) + log (X A [k ) + log (X A [k] n+1 ) + log (X A [*] ,
  • the spectral magnitude XA at index k of the current frame n may be modified as ⁇ to a magnitude corresponding to a geometric average of the spectral magnitude XA at index k of the preceding frame n-2, the spectral magnitude XA at index k of the preceding frame n-1 , the spectral magnitude XA at index k of the subsequent frame n+1 , and the spectral magnitude XA at index k of the subsequent frame n+2.
  • Equation 16 corresponds to an "averaging" function, which includes arguments consistent with the number of frames preceding or subsequent to the current frame n that have been analyzed (see operation 104) to determine whether a signal anomaly is present in the current frame n.
  • Equation 16 specifies the frames n-1 , n-2, n+1, and n+2, such as described above in connection with Equations 7-9.
  • particular spectral components of a current frame n may or may not be filtered (e.g., Equations 10-16) based on a comparison of respective spectral amplitude magnitudes at a particular index value k (e.g., Equations 4-9).
  • such an implementation may be performed for each respective frame n of the frequency domain signal X[ ] tract.
  • the frequency domain signal X'[ ] increment may be transformed into the time-domain according to an example N-point windowed IDFT (Inverse Discrete Fourier Transform) function of the form:
  • d[/] practice may correspond to the digitized form of the repaired (or not) input audio signal x(i), where the relationship between d[/] planet and an output accumulation buffer z[/] tract, where z[/] of the initial value n is zero (0), satisfies:
  • an output audio signal x'(i) may be generated by transforming the frequency domain signal X' [k]ograph into the time-domain.
  • an unfiltered segment 202 of an audio stream sample 204 prior to signal noise reduction, along with a filtered segment 206 of the audio stream sample 204 is shown in accordance with the present disclosure.
  • an amplitude A of the unfiltered segment 202 of the audio stream sample 204 is shown during an example time period t ⁇ 2.5 seconds.
  • an amplitude B of the filtered segment 206 of the audio stream sample 204 is shown during the example time period t ⁇ 2.5 seconds.
  • a particular signal anomaly 208 of the unfiltered segment 202 is shown within a "current" timeframe (n) 210, which is associated with the "current" frame n discussed above in connection with FIG. 1.
  • first "preceding" timeframe (n-l) 212 is associated with the frame n-l discussed above in connection with FIG. 1
  • first "subsequent" timeframe (n+l) 214 is associated with the frame n+l discussed above in connection with FIG. 1.
  • first "preceding" timeframe (n-l) 212 Immediately adjacent the first "preceding" timeframe (n-l) 212 is a second "preceding" timeframe (n-2) 216, and immediately adjacent the first "subsequent" timeframe (n+l) 214 is a second "subsequent" timeframe (n+2) 218.
  • the second "preceding" timeframe (n-2) 216 is associated with the frame n-2 discussed above in connection with FIG. 1
  • the second "subsequent" timeframe (n+2) 218 is associated with the frame n+2 discussed above in connection with FIG. 1.
  • the second "preceding" timeframe (n-2) 216 is defined by an interval dT between a time ti and a time t 2 .
  • the first "preceding" timeframe (n-l) 212 is defined by the interval dT between the time t 2 and a time t3.
  • the "current" timeframe (n) 210 is defined by then interval dT between the time t3 and a time t 4 .
  • the first "subsequent" timeframe (n+l) 214 is defined by the interval dT between the time t 4 and a time ts.
  • the second "subsequent" timeframe (n+2) 218 is defined by the interval dT between the time ts and a time t 6 .
  • the particular signal anomaly 208 within the "current" timeframe (n) 210 of the unfiltered segment 202 of the audio stream sample 204 may be identified in a manner such as described above in connection with FIG. 1.
  • the particular signal anomaly 208 is shown between the time t3 and the time t 4 of the unfiltered segment 202 of the audio stream sample 204.
  • the particular signal anomaly 208 is absent, having been removed or filtered, between the time t3 and the time t 4 of the filtered segment 206 of the audio stream sample 204.
  • only the frequency spectra of the first "preceding" timeframe (n-l) 212 and the frequency spectra of the first "subsequent" timeframe (n+l) 214 may be analyzed to identify and filter the particular signal anomaly 208.
  • the frequency spectra of the first "preceding" timeframe (n-l) 212 and the frequency spectra of the first "subsequent" timeframe (n+l) 214, along with the frequency spectra of the second "preceding" timeframe (n-2) 216 and the second "subsequent" timeframe (n+2) 218, may be analyzed to identify and filter the particular signal anomaly 208. Still other embodiments are possible.
  • the audio recording stage 302 includes a recording signal conditioning module 306, an A/D module 308, a recording noise reduction module 310, and a recording storage module 312.
  • the audio playback stage 304 includes a playback storage module 314, a playback noise reduction module 316, a D/A module 318, and a playback signal conditioning module 320.
  • Other embodiments of the audio recording stage 302 and the audio playback stage 304 are possible.
  • the audio recording stage 302 and the audio playback stage 304 may both be incorporated within a particular computing system or device (e.g., a server computer, laptop computer, smartphone, music player, etc.).
  • a particular computing system or device e.g., a server computer, laptop computer, smartphone, music player, etc.
  • one or more or the respective modules of the audio recording stage 302 and the audio playback stage 304 may be integral with, or combined with, one or more other ones of the respective modules of the audio recording stage 302 and the audio playback stage 304.
  • the A/D module 308 and the D/A module 318, and/or the recording signal conditioning module 306 and the playback signal conditioning module 320, and/or the recording noise reduction module 310 and the playback noise reduction module 316, and/or the recording storage module 312 and the playback storage module 314 may, respectively, be incorporated into a particular module implemented wholly or in part in hardware, software, or any combination thereof. Still other embodiment are possible.
  • the audio recording stage 302 and the audio playback stage 304 may include more or fewer modules as desired, and such a modification may or may not be implementation-specific.
  • the recording signal conditioning module 306 of the audio recording stage 302, or the playback signal conditioning module 320 of the audio playback stage 304 may be omitted.
  • respective modules of the audio recording stage 302 and/or the audio playback stage 304 may be rearranged, as desired, and such a modification may or may not be implementation-specific.
  • the playback noise reduction module 316, the D/A module 318, and the playback signal conditioning module 320 are shown coupled in series in direction of signal flow.
  • the input audio signal x(i) as discussed above in connection with FIG. 1 may be initially supplied to the recording signal conditioning module 306 for particular conditioning as desired such as, for example, amplification, filtering, converting, range matching, isolation, and others.
  • the conditioned signal may then be passed to the A/D module 308 for conversion into a digitized version of the input audio signal x(i).
  • the digitized version of the input audio signal x(i) may then be passed to the recording noise reduction module 310.
  • particular signal anomaly(ies) within the digitized version of the input audio signal x(i) may be identified and filtered in a manner such as described above in connection with FIGS. 1-2. Following such signal noise identification and reduction, a filtered, digital version of the input audio signal x(i) may be supplied to the recording storage module 312 for storage therein.
  • the input audio signal x(i) may be digitized and stored within a particular memory location without prior signal noise identification and reduction as discussed throughout the present disclosure.
  • a "post-storage" signal noise reduction may be implemented, followed by transfer of the respective noise-filtered signal for further processing, or use, as desired.
  • a digital input audio signal X(i) may be initially supplied to the playback noise reduction module 316 from the playback storage module 314.
  • particular signal anomaly(ies) within the digital input audio signal X(i) may be identified and filtered in a manner such as described above in connection with FIGS. 1-2.
  • the conditioned signal may then be passed to the D/A module 318 for conversion into an analog version of the input audio signal X(i).
  • the analog version of the input audio signal X(i) may then be passed to the playback signal conditioning module 320 for particular conditioning as desired.
  • Example scenarios or applications may include, among many others, simultaneous or near-simultaneous signal sharing or transmission applications (e.g., teleconferencing/videoconferencing), as well as “delayed” signal sharing or transmission applications (e.g., signal recording/playback).
  • simultaneous or near-simultaneous signal sharing or transmission applications e.g., teleconferencing/videoconferencing
  • “delayed” signal sharing or transmission applications e.g., signal recording/playback
  • FIG. 4 illustrates an embodiment of an example computer system/device 400.
  • An example of a computer device includes a mobile user equipment or terminal (e.g., smartphone), a server computer, desktop computer, laptop computer, personal data assistant, gaming console, and others.
  • the example computer device 400 may be configured to perform and/or include instructions that, when executed, cause the computer system 400 to perform the example method of FIG. 1.
  • FIG. 4 is intended only to provide a generalized illustration of various components, any or all of which may be utilized as appropriate. FIG. 4, therefore, broadly illustrates how individual system elements may be implemented in a relatively separated or relatively more integrated manner.
  • the computer device 400 is shown comprising hardware elements that can be electrically coupled via a bus 402 (or may otherwise be in communication, as appropriate).
  • the hardware elements may include a processing unit with one or more processors 404, including without limitation one or more general-purpose processors and/or one or more special-purpose processors (such as digital signal processing chips, graphics acceleration processors, and/or the like); one or more input devices 406, which can include without limitation a remote control, a mouse, a keyboard, and/or the like; and one or more output devices 408, which can include without limitation a presentation device (e.g., television), a printer, and/or the like.
  • a presentation device e.g., television
  • the computer system 400 may further include (and/or be in communication with) one or more non-transitory storage devices 410, which can comprise, without limitation, local and/or network accessible storage, and/or can include, without limitation, a disk drive, a drive array, an optical storage device, a solid-state storage device, such as a random access memory (“RAM”), and/or a read-only memory (“ROM”), which can be programmable, flash-updateable, and/or the like.
  • RAM random access memory
  • ROM read-only memory
  • Such storage devices may be configured to implement any appropriate data stores, including without limitation, various file systems, database structures, and/or the like.
  • the computer device 400 might also include a communications subsystem 412, which can include without limitation a modem, a network card (wireless or wired), an infrared communication device, a wireless communication device, and/or a chipset (such as a BluetoothTM device, an 402.11 device, a WiFi device, a WiMax device, cellular communication facilities (e.g., GSM, WCDMA, LTE, etc.), and/or the like.
  • the communications subsystem 412 may permit data to be exchanged with a network (such as the network described below, to name one example), other computer systems, and/or any other devices described herein.
  • one or more procedures described with respect to the method(s) discussed above, and/or system components might be implemented as code and/or instructions executable by a computer (and/or a processor within a computer); in an aspect, then, such code and/or instructions can be used to configure and/or adapt a general purpose computer (or other device) to perform one or more operations in accordance with the described methods.
  • a set of these instructions and/or code might be stored on a non-transitory computer- readable storage medium, such as the storage device(s) 410 described above.
  • the storage medium may be incorporated within a computer system, such as computer system 400.
  • the storage medium might be separate from a computer system (e.g., a removable medium, such as flash memory), and/or provided in an installation package, such that the storage medium can be used to program, configure, and/or adapt a general purpose computer with the instructions/code stored thereon.
  • These instructions might take the form of executable code, which is executable by the computer device 400 and/or might take the form of source and/or installable code, which, upon compilation and/or installation on the computer system 400 (e.g., using any of a variety of generally available compilers, installation programs, compression/decompression utilities, etc.), then takes the form of executable code.
  • some embodiments may employ a computer system (such as the computer device 400) to perform methods in accordance with various embodiments of the invention. According to a set of embodiments, some or all of the procedures of such methods are performed by the computer system 400 in response to processor 404 executing one or more sequences of one or more instructions (which might be incorporated into the operating system 416 and/or other code, such as an application program 418) contained in the working memory 414. Such instructions may be read into the working memory 414 from another computer-readable medium, such as one or more of the storage device(s) 410. Merely by way of example, execution of the sequences of instructions contained in the working memory 414 might cause the processor(s) 404 to perform one or more procedures of the methods described herein.
  • a computer system such as the computer device 400
  • machine-readable medium and “computer-readable medium,” as used herein, refer to any medium that participates in providing data that causes a machine to operate in a specific fashion.
  • various computer-readable media might be involved in providing instructions/code to processor(s) 404 for execution and/or might be used to store and/or carry such instructions/code.
  • a computer-readable medium is a physical and/or tangible storage medium.
  • Such a medium may take the form of a non-volatile media or volatile media.
  • Non-volatile media include, for example, optical and/or magnetic disks, such as the storage device(s) 410.
  • Volatile media include, without limitation, dynamic memory, such as the working memory 414.
  • Common forms of physical and/or tangible computer-readable media include, for example, a floppy disk, a flexible disk, hard disk, magnetic tape, or any other magnetic medium, a CD-ROM, any other optical medium, punchcards, papertape, any other physical medium with patterns of holes, a RAM, a PROM, EPROM, a FLASH-EPROM, any other memory chip or cartridge, or any other medium from which a computer can read instructions and/or code.
  • Various forms of computer-readable media may be involved in carrying one or more sequences of one or more instructions to the processor(s) 404 for execution.
  • the instructions may initially be carried on a magnetic disk and/or optical disc of a remote computer.
  • a remote computer might load the instructions into its dynamic memory and send the instructions as signals over a transmission medium to be received and/or executed by the computer system 400.
  • the communications subsystem 412 (and/or components thereof) generally will receive signals, and the bus 402 then might carry the signals (and/or the data, instructions, etc. carried by the signals) to the working memory 414, from which the processor(s) 404 retrieves and executes the instructions.
  • the instructions received by the working memory 414 may optionally be stored on a non-transitory storage device 410 either before or after execution by the processor(s) 404.
  • the methods, systems, and devices discussed above are examples. Various configurations may omit, substitute, or add various method steps or procedures, or system components as appropriate. For instance, in alternative configurations, the methods may be performed in an order different from that described, and/or various stages may be added, omitted, and/or combined. Also, features described with respect to certain configurations may be combined in various other configurations. Different aspects and elements of the configurations may be combined in a similar manner. Also, technology evolves and, thus, many of the elements are examples and do not limit the scope of the disclosure or claims.
  • configurations may be described as a process which is depicted as a flow diagram or block diagram. Although each may describe the operations as a sequential process, many of the operations can be performed in parallel or concurrently. In addition, the order of the operations may be rearranged. A process may have additional steps not included in the figure.
  • examples of the methods may be implemented by hardware, software, firmware, middleware, microcode, hardware description languages, or any combination thereof. When implemented in software, firmware, middleware, or microcode, the program code or code segments to perform the necessary tasks may be stored in a non-transitory computer-readable medium such as a storage medium. Processors may perform the described tasks.
  • the example embodiments described herein may be implemented as logical operations in a computing device in a networked computing system environment.
  • the logical operations may be implemented as any combination of: (i) a sequence of computer implemented instructions, steps, or program modules running on a computing device; and (ii) interconnected logic or hardware modules running within a computing device.

Abstract

L'invention concerne des systèmes et des procédés pour une réduction de bruit de signal. Un signal numérique d'entrée peut être séparé en une série de segments adjacents. Les segments adjacents peuvent être convertis en une représentation de domaine de fréquence. Une composante spectrale particulière d'un segment particulier peut être comparé à une composante spectrale associée d'un premier segment adjacent au segment particulier, et à une composante spectrale associée d'un second segment adjacent au segment particulier. La composante spectrale particulière peut être modifiée lorsqu'une valeur d'amplitude de la composante spectrale particulière satisfait au moins une condition d'un ensemble prédéterminé de conditions.
PCT/CN2013/089189 2013-12-12 2013-12-12 Réduction de bruit de signal WO2015085532A1 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/CN2013/089189 WO2015085532A1 (fr) 2013-12-12 2013-12-12 Réduction de bruit de signal

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2013/089189 WO2015085532A1 (fr) 2013-12-12 2013-12-12 Réduction de bruit de signal

Publications (1)

Publication Number Publication Date
WO2015085532A1 true WO2015085532A1 (fr) 2015-06-18

Family

ID=53370491

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2013/089189 WO2015085532A1 (fr) 2013-12-12 2013-12-12 Réduction de bruit de signal

Country Status (1)

Country Link
WO (1) WO2015085532A1 (fr)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105913855A (zh) * 2016-04-11 2016-08-31 宁波大学 一种基于长窗比例因子的回放语音攻击检测算法
CN106098076A (zh) * 2016-06-06 2016-11-09 成都启英泰伦科技有限公司 一种基于动态噪声估计时频域自适应语音检测方法

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050058301A1 (en) * 2003-09-12 2005-03-17 Spatializer Audio Laboratories, Inc. Noise reduction system
CN101038744A (zh) * 2005-11-28 2007-09-19 索尼株式会社 音频信号噪声降低设备和方法
CN101114451A (zh) * 2006-07-27 2008-01-30 奇景光电股份有限公司 噪声消除系统及其数字音频处理单元
CN101944364A (zh) * 2009-07-09 2011-01-12 展讯通信（上海）有限公司 音频处理方法及音频系统

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050058301A1 (en) * 2003-09-12 2005-03-17 Spatializer Audio Laboratories, Inc. Noise reduction system
CN101038744A (zh) * 2005-11-28 2007-09-19 索尼株式会社 音频信号噪声降低设备和方法
CN101114451A (zh) * 2006-07-27 2008-01-30 奇景光电股份有限公司 噪声消除系统及其数字音频处理单元
CN101944364A (zh) * 2009-07-09 2011-01-12 展讯通信（上海）有限公司 音频处理方法及音频系统

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105913855A (zh) * 2016-04-11 2016-08-31 宁波大学 一种基于长窗比例因子的回放语音攻击检测算法
CN106098076A (zh) * 2016-06-06 2016-11-09 成都启英泰伦科技有限公司 一种基于动态噪声估计时频域自适应语音检测方法

Similar Documents

Publication Publication Date Title
US8559568B1 (en) Sliding DFT windowing techniques for monotonically decreasing spectral leakage
US9343075B2 (en) Voice processing apparatus and voice processing method
EP3364413A1 (fr) Procédé de détermination de signal de bruit, et procédé et dispositif destinés à la suppression de bruit audio
US9767829B2 (en) Speech signal processing apparatus and method for enhancing speech intelligibility
US20150179184A1 (en) Compensating For Identifiable Background Content In A Speech Recognition Device
US20130338806A1 (en) System and method for selective removal of audio content from a mixed audio recording
KR20190102956A (ko) 혼합 뉴머롤로지 시스템에서 데이터의 간섭을 제거하는 방법 및 장치
US9299361B2 (en) Signal noise reduction
CN112309426A (zh) 语音处理模型训练方法及装置和语音处理方法及装置
CN111863008A (zh) 一种音频降噪方法、装置及存储介质
WO2015085532A1 (fr) Réduction de bruit de signal
CN110062945B (zh) 音频输入信号的处理
US10667055B2 (en) Separated audio analysis and processing
US9495978B2 (en) Method and device for processing a sound signal
CN112002339B (zh) 语音降噪方法和装置、计算机可读的存储介质及电子装置
US10157049B2 (en) Static analysis with input reduction
CN114255785A (zh) 音频检测方法、装置、终端及存储介质
CN113613112A (zh) 抑制麦克风的风噪的方法和电子装置
Kehtarnavaz et al. Using smartphones as mobile implementation platforms for applied digital signal processing courses
CN112067927A (zh) 中高频振荡检测方法及装置
CN110851258A (zh) Api调用方法、装置、计算机设备及存储介质
US10044846B2 (en) Method for executing dual operating systems of smart phone
Verma et al. Pre-processing
US10360026B1 (en) System, method, and computer program for RPM versioning
US20170337935A1 (en) Speech processing apparatus, speech processing method, and recording medium

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 13899193

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 13899193

Country of ref document: EP

Kind code of ref document: A1