Noise Update Circuit

Abstract

To provide a noise update circuit that accurately subtracts noise components from a voice signal to be input, and is used in a noise reduction circuit.SOLUTION: A noise update circuit has: an average value calculation unit that calculates an average spectra level M for an input signal spectrum Yi(f) corresponding to an amplitude spectrum of a frame to be processed; a first update unit that calculates a post-update noise spectrum Ni(f) for a frequency f where Yi(f)<Te×M regarding the input signal spectrum Yi(f) and the average spectra level M when the frame to be processed is a frame including only noise components; a second update unit that determines a post-update noise spectrum Ni(f) for a frequency where Yi(f)≥Te×M regarding the input signal spectrum Yi(f) and the average spectra level M; and a third update unit that determines a post-update noise spectrum Ni(f) for each frequency f when the frame to be processed is a frame including voice components.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present invention relates to a noise updating circuit, and more particularly to a noise updating circuit that can be used in a noise reduction circuit incorporated in, for example, a wireless device that transmits and receives high-frequency signals. [Background technology]

[0002] Spectral subtraction is known as a method for suppressing noise components contained in an audio signal (see, for example, Patent Document 1 and Non-Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 4-238399 [Non-patent literature]

[0004] [Non-Patent Document 1] P. Scalart and J. Vieira Filho, "Speech Enhancement Based on a Priori Signal to Noise Estimation", IEEE International Conference on. Acoustics, Speech, Signal Processing, Atlanta, GA, USA, vol. 2, pp. 629-632, 1996. Summary of the Invention [Problem to be solved by the invention]

[0005] Incidentally, in order to appropriately apply the spectral subtraction method, it is important to accurately subtract noise components from the input audio signal, if necessary, taking into account the needs of the user.

[0006] SUMMARY OF THE PRESENT DISCLOSURE An object of the present invention is to provide a noise updating circuit that is capable of accurately subtracting noise components from an input audio signal, taking into account the needs of the user as necessary. [Means for solving the problem]

[0007] In order to solve the above problem, the present invention provides a signal processing method and apparatus comprising: an average calculation unit that calculates an average spectrum level M for an input signal spectrum Yi(f) of a signal corresponding to an amplitude spectrum of a frame to be processed, in accordance with the following formula 1 (where f: frequency in the amplitude spectrum, f1: minimum frequency in the amplitude spectrum, f2: maximum frequency in the amplitude spectrum, Fn: number of frequencies in the range from the minimum frequency f1 to the maximum frequency f2, and i: ordinal number representing a time series order);

number

number

number

number

number

[0009] Claim 2 The invention described in claim to 1 The noise updating circuit described in the above specification is characterized in that the coefficient Te is set to any value within the range of 1 to 100. Effect of the Invention

[0010] Claim to 1 According to the invention described in claim 1, it is possible to avoid suppressing tone signals. Specifically, in a conventional circuit that realizes a spectral subtraction method, components that are constantly present in a frequency spectrum are judged as noise and suppressed, so tone signals are also judged as noise due to their stationary nature and are subject to suppression, resulting in the problem that tone signals necessary for users (e.g., Morse code) are also suppressed. In contrast, in the invention described in claim 1, frequency components whose amplitude spectrum is significantly larger than the average spectrum level are excluded from updating the noise spectrum, making it possible to avoid suppressing tone signals.

[0011] Claim 2According to the invention described in the above, the coefficient Te is set to an appropriate value, so that it is possible to more reliably exclude frequency components whose amplitude spectrum is significantly larger than the average spectrum level from the updating of the noise spectrum, and it is possible to more reliably prevent tone signals from being suppressed. [Brief description of the drawings]

[0012]

Figure 1

Figure 2

[0013] Hereinafter, the present invention will be described based on the illustrated embodiment.

[0014] Fig. 1 is a functional block diagram showing a schematic configuration of a noise reduction circuit 1 including a noise updating circuit 11 according to an embodiment of the present invention. Fig. 2 is a functional block diagram showing a schematic configuration of the noise updating circuit 11 according to the embodiment.

[0015] The noise reduction circuit 1 is a circuit that is incorporated, for example, in a radio device that transmits and receives high-frequency signals, and realizes Spectral Subtraction, which is a method for suppressing noise components contained in an audio signal. The noise reduction circuit 1 mainly includes a pre-emphasis circuit 2, a window processing unit 3, a time-frequency conversion unit 4, a conversion result output unit 5, a subtraction unit 6, a synthesis unit 7, a frequency-time conversion unit 8, a de-emphasis circuit 9, a voice activity detection unit 10, and a noise updating circuit 11.

[0016] A pre-emphasis (PE) circuit 2 performs high-frequency emphasis processing for amplifying the relative intensity of high-frequency components in advance on an audio signal obtained by demodulating a high-frequency signal received from an antenna, and outputs the signal after high-frequency emphasis processing.

[0017] The window processing unit 3 receives the high-frequency emphasis processed signal output from the pre-emphasis circuit 2, extracts frames of a predetermined length from the input signal (e.g., extracts a time waveform of 25 ms every 12.5 ms), and performs window processing on each frame by multiplying it by a window function such as a Hanning window. The window processing unit 3 outputs the windowed frame each time it performs window processing on each frame.

[0018] The time-frequency conversion unit 4 receives a windowed frame output from the window processing unit 3, and each time the time-frequency conversion unit 4 receives a frame, it performs a conversion process from a time domain signal to a frequency domain signal, calculates a frequency spectrum including amplitude components and phase components for each of a plurality of frequencies, and outputs a signal of a frequency spectrum of real and imaginary numbers. The time-frequency conversion unit 4 performs a time-frequency conversion by, for example, a Discrete Fourier Transform or a Fast Fourier Transform to calculate the frequency spectrum.

[0019] The conversion result output unit 5 receives an input of a frequency spectrum signal for each frame (for example, at intervals of about 12.5 ms) output from the time-frequency conversion unit 4, and outputs, for each frame, a signal corresponding to an amplitude spectrum including amplitude components of each frequency in the input frequency spectrum to a subtraction unit 6, and outputs a signal corresponding to a phase spectrum including phase components of each frequency in the input frequency spectrum to a synthesis unit 7.

[0020] The subtraction unit 6 receives an input of a signal corresponding to the amplitude spectrum for each frame output from the conversion result output unit 5, and also receives an input of a signal corresponding to the updated noise spectrum for each frame output from the noise update circuit 11, and subtracts the signal corresponding to the input updated noise spectrum for each frequency (in other words, for each spectrum) from the signal corresponding to the input amplitude spectrum for each frame. This suppresses the noise components contained in the audio signal. The subtraction unit 6 outputs a signal corresponding to the amplitude spectrum after subtraction for each frame output from the conversion result output unit 5.

[0021] The synthesis unit 7 receives an input of a signal corresponding to the phase spectrum for each frame output from the conversion result output unit 5, and also receives an input of a signal corresponding to the amplitude spectrum after the subtraction process for each frame output from the subtraction unit 6, synthesizes the input signal corresponding to the phase spectrum and the signal corresponding to the amplitude spectrum for each frame to generate a frequency spectrum, and outputs a signal with a frequency spectrum of real and imaginary numbers.

[0022] The frequency-time transform unit 8 receives the frequency spectrum signal for each frame output from the synthesis unit 7, and performs a transform process from a frequency domain signal to a time domain signal for each frame on the input frequency spectrum signal, i.e., an inverse transform process of the transform process in the time-frequency transform unit 4, to output an audio signal. The frequency-time transform unit 8 performs frequency-time transform, for example, by inverse discrete Fourier transform or inverse fast Fourier transform, to generate an audio signal.

[0023] The de-emphasis (DE) circuit 9 receives the audio signal output from the frequency-time conversion unit 8, performs high-frequency attenuation processing on the input audio signal to attenuate the relative intensity of high-frequency components, i.e., attenuation processing using the inverse filter of the pre-emphasis circuit 2, and outputs the audio signal after high-frequency attenuation processing.

[0024] The voice activity detection unit 10 receives an input of a signal corresponding to the amplitude spectrum for each frame output and branched from the conversion result output unit 5, and judges for each frame, with respect to the input signal corresponding to the amplitude spectrum, whether it is a frame containing only noise components or a frame including voice components.

[0025] The method used by the voice activity detection unit 10 to determine whether a frame to be processed contains only noise components or also contains voice components is not limited to a specific procedure or technique, but may be appropriately selected from conventional or new procedures or techniques.

[0026] As a method for determining whether a frame to be processed contains only noise components or also contains voice components in the voice section detection unit 10, for example, a method that focuses on the non-constancy of voice and determines that the frame to be processed contains only noise components if the average or variance value of the amplitude magnitude for each frequency of the amplitude spectrum is less than a predetermined threshold value multiple times (e.g., about 3 to 5 times) consecutively in the most recent frame, and determines that the frame to be processed contains voice components in other cases; or a method that determines that the frame to be processed contains only noise components if the average or variance value of the amplitude magnitude for each frequency of the amplitude spectrum is less than a predetermined threshold value, and determines that the frame to be processed contains voice components if the average or variance value is equal to or greater than the threshold value.

[0027] The voice activity detection unit 10 outputs a noise frame signal when it determines that the frame to be processed contains only noise components, and outputs a voice frame signal when it determines that the frame to be processed contains voice components. The voice activity detection unit 10 outputs a noise frame signal or a voice frame signal as the voice activity detection result for each frame.

[0028] Then, the noise update circuit 11 according to the embodiment takes a signal corresponding to the amplitude spectrum of the frame to be processed as the input signal spectrum Yi(f), and has an average value calculation unit 111 that calculates the average spectrum level M for the input signal spectrum Yi(f) (where f is the frequency in the amplitude spectrum and i is the ordinal number representing the time series order). When the frame to be processed is a frame consisting only of noise components, for the frequency f where Yi(f) < Te × M with respect to the input signal spectrum Yi(f) and the average spectrum level M, taking the signal corresponding to the amplitude spectrum of the frame to be processed as the input signal spectrum Yi(f), a first update unit 112 that calculates the updated noise spectrum Ni(f) for each frequency f by a process including averaging (where Te is a coefficient). When the frame to be processed is a frame consisting only of noise components, for the frequency f where Yi(f) ≥ Te × M with respect to the input signal spectrum Yi(f) and the average spectrum level M, a second update unit 113 that determines the updated noise spectrum Ni(f) for each frequency f. When the frame to be processed is a frame including a voice component, a third update unit 114 that determines the updated noise spectrum Ni(f) for each frequency f, and is configured in this way.

[0029] The noise update circuit 11 updates the latest noise spectrum by taking into account the amplitude spectrum of the current frame (in other words, the frame to be processed, the latest frame) in the noise spectrum representing the noise components for each frequency calculated in the past, and has an average value calculation unit 111, a first update unit 112, a second update unit 113, and a third update unit 114.

[0030] The noise updating circuit 11 receives an input of a signal corresponding to the amplitude spectrum for each frame output and branched from the conversion result output unit 5, as well as an input of the voice activity detection result for each frame output from the voice activity detection unit 10, and updates the noise spectrum Ni-1(f) of the previous frame for each frequency f using the input signal corresponding to the amplitude spectrum, and calculates the updated noise spectrum Ni(f) according to one of the following formulas 7 to 10 depending on the content of the input voice activity detection result, etc. Note that the subscript i in the following formulas is an ordinal number indicating the order in time series, and indicates an order commonly applied to all formulas.

[0031] The average value calculation unit 111 calculates an average spectrum level M according to the following formula 6, using the signal corresponding to the amplitude spectrum for each frame output and branched from the conversion result output unit 5 as the input signal spectrum Yi(f). The average spectrum level M is the average value in the frequency axis direction of the amplitude spectrum.

number

[0032] In Equation 6, f represents a frequency in the amplitude spectrum, f1 is the minimum frequency in the amplitude spectrum, f2 is the maximum frequency in the amplitude spectrum, and Fn represents the number of frequencies in the range from the minimum frequency f1 to the maximum frequency f2. In other words, the number of frequencies Fn is n / 2+1, where n is the number of sample points included in one frame in the time-frequency conversion by the time-frequency conversion unit 4.

[0033] The average value calculation section 111 calculates an average spectrum level M every time a signal corresponding to the amplitude spectrum of each frame is input.

[0034] Furthermore, when the voice section detection result input to the noise update circuit 11 is a noise frame signal, for each frequency f, the signal corresponding to the input amplitude spectrum is used as the input signal spectrum Yi(f), and the following process or is performed according to the relationship between the input signal spectrum Yi(f) and the average spectrum level M.

[0035] For the frequency f where Yi(f) < Te × M The first update unit 112 calculates the updated noise spectrum Ni(f) for each frequency f according to the following formula 7 which is an IIR (Infinite Impulse Response; infinite impulse response) filter or the following formula 8 which is an FIR (Finite Impulse Response; finite impulse response) filter. Note that Ni-1(f) in the following formulas represents the noise spectrum one frame before the update. Also, the subscript j in formula 8 is a variable representing the degree of separation from the ordinal number i in the time series, and is an integer of 0 or more. Furthermore, Yi-j(f) in formula 8 represents the input signal spectrum j frames before the update. [Number] [Number]

[0036] In the relationship between the input signal spectrum Yi(f) and the average spectrum level M, Te is a coefficient for excluding frequency components whose amplitude spectrum is significantly larger than the average spectrum level M from the update of the noise spectrum. The coefficient Te is not limited to a specific value. For example, considering that frequency components corresponding to tone signals are excluded from the update of the noise spectrum so that the tone signals are not subtracted by the subtraction unit 6 and are not suppressed as noise components, it is appropriately set to an appropriate value. Specifically, the coefficient Te is considered to be set to any value in the range of about 1 to 100, and particularly considered to be set to about 10.

[0037] In Equation 8, K is a constant that determines the weighting of the noise spectrum Ni-1(f) of the frame before the update to the input signal spectrum Yi(f), which is the amplitude spectrum of the frame to be processed (in other words, the current frame, the latest frame), and is called the “pre-update weighting constant K.”

[0038] The pre-update weighting constant K is not limited to a specific value as long as it is an integer equal to or greater than 0. Specifically, the pre-update weighting constant K may be set to any value within a range equivalent to about 0.06 to 0.20 seconds of the time constant of the IIR filter or the average interval of the FIR filter (for example, a range of about K=5 to 16 at a frame interval of 12.5 ms), and may be set to a value equivalent to about 0.1 seconds of the time constant of the IIR filter or the average interval of the FIR filter (for example, about K=8 at a frame interval of 12.5 ms). Note that when the time constant of the IIR filter or the average interval of the FIR filter is set to, for example, 0.1 seconds, the specific value of the constant K in Equation 7 is different from the specific value of the constant K in Equation 8.

[0039] In addition, in the process including averaging for calculating the updated noise spectrum Ni(f), when an IIR filter (general form is Ni(f)=ΣAjYi-j(f)+ΣBkNi-k(f); where A and B are coefficients, i is an ordinal number representing the order of the time series, j is an integer of 0 or more representing the degree of distance from the ordinal number i in the time series, and k is an integer of 1 or more representing the degree of distance from the ordinal number i in the time series) is used, an averaging time of the average using the input signal spectrum Yi(f), the past input signal spectrum Yi-j(f), and the past noise spectrum Ni-k(f) may be used when calculating the updated noise spectrum Ni(f).

[0040] In addition, when an IIR filter with Aj=0 for j≧1 is used for the general form of the IIR filter described above, an averaging time for averaging the input signal spectrum Yi(f) and the past noise spectrum Ni-k(f) may be used when calculating the updated noise spectrum Ni(f).

[0041] In addition, for the general form of the IIR filter above, when Aj=0 for j≧1, A0=1 / (1+K), B1=K / (1+K), and Bk=0 (k≧2), the above formula 7 applies.

[0042] Furthermore, in the process including averaging for calculating the updated noise spectrum Ni(f), when an FIR filter (whose general form is Ni(f)=ΣAjYi-j(f); where A is a coefficient, i is an ordinal number representing the order of the time series, and j is an integer equal to or greater than 0 representing the degree of distance from the ordinal number i in the time series; in other words, the general form of the IIR filter described above with Bk=0) is used, an averaging time for the average using the input signal spectrum Yi(f) and the past input signal spectrum Yi-j(f) may be used when calculating the updated noise spectrum Ni(f).

[0043] In addition, regarding the general form of the above FIR filter, the case where Aj=1 / K corresponds to the above formula 8.

[0044] For frequency f where Yi(f) ≧ Te × M The second update unit 113 determines the updated noise spectrum Ni(f) according to the following Equation 9.

number

[0045] In Equation 9, Ni-1(f) represents the noise spectrum of the frame before the update.

[0046] That is, for frequency components where Yi(f)≧Te×M and the amplitude spectrum is significantly greater than the average spectrum level M, the noise spectrum Ni(f) after the update is left as the noise spectrum Ni-1(f) of one frame before the update and is excluded from the update.

[0047] Furthermore, when the voice section detection result input to the noise updating circuit 11 is a voice frame signal, the third updating unit 114 determines the updated noise spectrum Ni(f) for each frequency f according to the following Equation 10.

number

[0048] Every time the noise updating circuit 11 receives the signal corresponding to the amplitude spectrum for each frame and the voice activity detection result, it outputs a signal corresponding to the updated noise spectrum Ni(f) for each frequency f to the subtraction unit 6. The subtraction unit 6 performs a process of subtracting the signal corresponding to the updated noise spectrum Ni(f) from the signal corresponding to the amplitude spectrum output from the conversion result output unit 5, using the signal corresponding to the updated noise spectrum Ni(f) output from the noise updating circuit 11 for each frame.

[0049] The above-described noise updating circuit 11 makes it possible not to suppress tone signals. Specifically, in conventional circuits that realize the spectral subtraction method, components that are constantly present in the frequency spectrum are judged to be noise and suppressed, so tone signals are also judged to be noise due to their stationary nature and are subject to suppression, resulting in the problem that tone signals necessary for users (e.g., Morse code) are also suppressed. In contrast, the above-described noise updating circuit 11 excludes frequency components whose amplitude spectrum is significantly larger than the average spectrum level from the noise spectrum update, making it possible not to suppress tone signals.

[0050] Although the embodiment of the present invention has been described above, the specific configuration is not limited to the above embodiment, and even if the design is changed within the scope of the present invention, it is included in the present invention. For example, in the above embodiment, the noise updating circuit 11 according to the present invention is applied to the noise reduction circuit 1 whose schematic configuration is shown in FIG. 1, but the configuration of the noise reduction circuit to which the present invention can be applied is not limited to the example shown in FIG. 1. As an example, the circuit configuration may be such that the user can select, as necessary, whether to suppress the noise component by using the circuit configuration specific to the present invention (particularly, the average value calculation unit 111 and the second updating unit 113) or to suppress the noise component in a conventional manner without using the circuit configuration specific to the present invention. In addition, the circuit to which the present invention can be applied is not limited to the noise reduction circuit. In other words, the present invention can be applied to various circuits in which it is necessary to update the noise spectrum in a time series. [Explanation of symbols]

[0051] 1 Noise reduction circuit 2 Pre-emphasis circuit 3 Window processing section 4 Time-frequency transform section 5. Conversion result output section 6 Subtraction section 7. Synthesis section 8 Frequency-time conversion section 9 De-emphasis circuit 10 Voice Activity Detector 11 Noise Update Circuit 111 Average value calculation section 112 First Update 113 Second Update 114 Third Update

Claims

1. an average calculation unit that calculates an average spectrum level M for an input signal spectrum Yi(f) corresponding to an amplitude spectrum of a frame to be processed in accordance with the following Equation 1 (where f: frequency in the amplitude spectrum, f1: minimum frequency in the amplitude spectrum, f2: maximum frequency in the amplitude spectrum, Fn: number of frequencies in the range from the minimum frequency f1 to the maximum frequency f2, and i: ordinal number representing the order in time series); [0010] a first updating unit that calculates an updated noise spectrum Ni(f) for each frequency f where Yi(f)<Te×M with respect to the input signal spectrum Yi(f) and the average spectrum level M when the frame to be processed is a frame containing only noise components, according to the following Equation 2 which is an IIR filter or the following Equation 3 which is an FIR filter (where Ni-1(f): noise spectrum one frame before the update, Yi-j(f): input signal spectrum j frames before the update, K: constant for determining weighting of Ni-1(f) with respect to Yi(f), j: integer equal to or greater than 0 which indicates the degree of separation from ordinal number i in the time series); [0025] [0030] a second update unit that determines an updated noise spectrum Ni(f) for each frequency f where Yi(f)≧Te×M with respect to the input signal spectrum Yi(f) and the average spectrum level M when the frame to be processed is a frame containing only noise components, according to the following Equation 4 (where Te is a coefficient, Ni-1(f) is the noise spectrum one frame before the update); [0045] a third update unit that determines an updated noise spectrum N(f) for each frequency f according to Equation 5 below when the frame to be processed is a frame including a speech component; [0050] A noise updating circuit comprising:

2. The coefficient Te is set to any value in the range of 1 to 100.

2. The noise updating circuit according to claim 1 .

Images