CN108366331B

CN108366331B - Audio processing device and audio processing method

Info

Publication number: CN108366331B
Application number: CN201710059978.9A
Authority: CN
Inventors: 谢沛彣
Original assignee: Realtek Semiconductor Corp
Current assignee: Realtek Semiconductor Corp
Priority date: 2017-01-24
Filing date: 2017-01-24
Publication date: 2020-10-02
Anticipated expiration: 2037-01-24
Also published as: CN108366331A

Abstract

The application provides an audio processing device and an audio processing method, wherein the audio processing device comprises an inverse noise filter, an output circuit and an equalizer circuit. The inverse noise filter is used for processing the digital signal to generate a noise elimination signal. The output circuit is used for mixing the noise elimination signal and the equalization signal to generate a mixing signal and generating an acoustic output signal based on the mixing signal, wherein the digital signal is related to the acoustic output signal. The equalizer circuit is used for receiving an input signal and adjusting at least one parameter in the equalizer circuit according to the equalized signal and the digital signal so as to process the input signal into an equalized signal.

Description

Audio processing device and audio processing method

Technical Field

The present disclosure relates to an audio processing apparatus, and more particularly, to an audio processing apparatus and method with detecting a state of an earphone.

Background

In order to provide higher sound quality, active noise cancellation mechanisms are often added to headphones to reduce the effects of ambient noise. In some techniques, the active noise cancellation mechanism is implemented by a feedback circuit. In these techniques, after an audio signal is output from a speaker, the audio signal is received by an internal microphone and fed back to the audio mixing device. In order to prevent the original audio from being affected by the fed back audio signal, an equalizer is required to be added to provide compensation.

Generally, the music signal fed back is changed by external factors (e.g., the position of the earphone, the shape of the user's ear). However, in the prior art, the equalizer can only provide a fixed transfer function for compensation. Therefore, when the music signal fed back changes, the equalizer cannot provide the corresponding sound effect.

Disclosure of Invention

To solve the above problem, in some embodiments, an audio processing apparatus includes an inverse noise filter, an output circuit, and an equalizer circuit. The inverse noise filter is used for processing the digital signal to generate a noise elimination signal. The output circuit is used for mixing the noise elimination signal and the equalization signal to generate a mixing signal and generating an acoustic output signal based on the mixing signal, wherein the digital signal is related to the acoustic output signal. The equalizer circuit is used for receiving an input signal and adjusting at least one parameter in the equalizer circuit according to the equalized signal and the digital signal so as to process the input signal into an equalized signal.

In some embodiments, the audio processing method of the present disclosure includes the following operations: processing the digital signal to generate a noise cancellation signal; mixing the noise cancellation signal and the equalization signal to generate a mixed signal, and outputting an audio output signal based on the mixed signal, wherein the digital signal is associated with the audio output signal; and adjusting at least one parameter according to the equalized signal and the digital signal to process the input signal into an equalized signal.

In summary, the audio processing apparatus and method provided by the present disclosure can detect the position of the earphone or different ear configurations to dynamically adjust the parameters of the equalizer to maintain the final output sound effect.

Drawings

In order to make the aforementioned and other objects, features, advantages and embodiments of the disclosure more comprehensible, the description of the drawings is as follows:

fig. 1 is a schematic diagram illustrating an audio processing device according to some embodiments of the present disclosure;

FIG. 2A is a flow chart of a method as performed by the control circuit of FIG. 1, shown in some embodiments according to the present disclosure;

FIG. 2B is a functional block diagram illustrating a method for performing the method of FIG. 2A according to some embodiments of the present disclosure;

FIG. 3 is a schematic diagram illustrating an audio processing device according to some embodiments of the present disclosure; and

fig. 4 is a flow diagram illustrating an audio processing method according to some embodiments of the present disclosure.

Description of reference numerals:

100. 300, and (2) 300: the audio processing apparatus 110: analog-to-digital converter

120: inverse noise filter 130: output circuit

140: the equalizer circuit 150: acoustic-electric conversion device

V (t): noise signal so (t): sound output signal

Y (n): digital signal e (t): electronic signal

Sinv (z), D (z): transfer functions h (z), s (z): transfer function

NC (n): noise cancellation signal 132: arithmetic circuit

134: digital-to-analog converter 136: electroacoustic transducer

141: the equalizer 142: adaptive filter

143: reverse noise filter 144: arithmetic circuit

142A: filter circuit 142B: control circuit

M (n): music signal u (n): mixed frequency signal

a (n): equalization signals F1(n), F2 (n): filtered signal

EO (n): equalization output signals S210 to S240: operation of

200. 400:

methods

210A, 220A: unit cell

S410 to S430:

operations

230A, 240A: unit cell

Detailed Description

The following detailed description is provided by way of example only and with reference to the accompanying drawings, which are not intended to limit the scope of the present disclosure, but rather are intended to limit the order of execution of the embodiments, and any arrangement of components or structures which results in a technically equivalent result is intended to be within the scope of the present disclosure. In addition, the drawings are for illustrative purposes only and are not drawn to scale. For ease of understanding, like elements in the following description will be described with like reference numerals.

As used herein, "first," "second," …, etc., are not intended to be limited to the exact meaning of sequence or order, nor are they intended to be limiting of the disclosure, but rather are intended to distinguish between elements or operations described in the same technical language. Further, as used herein, the term "couple" or "connect" refers to two or more elements being in direct physical or electrical contact with each other, or in indirect physical or electrical contact with each other, or to the mutual operation or action of two or more elements.

Fig. 1 is a schematic diagram illustrating an audio processing device 100 according to some embodiments of the present disclosure. In some embodiments, the audio processing device 100 may be disposed on a headset. In some embodiments, the audio processing apparatus 100 has an active noise cancellation mechanism to reduce the interference of the environmental noise.

In some embodiments, the audio processing apparatus 100 includes an analog-to-digital converter 110, an inverse noise filter 120, an output circuit 130, an equalizer circuit 140, and an acoustoelectric conversion apparatus 150.

In some embodiments, the sound-electricity converting device 150 is disposed in the housing of the earphone and is configured to receive the sound output signal so (t) and the noise signal v (t). The sound-to-electricity conversion device 150 converts the received sound audio signal into an electronic signal e (t). In some embodiments, the sound-electricity converting device 150 may be implemented by a microphone, but the disclosure is not limited thereto.

The analog-to-digital converter 110 is used for converting the electronic signal e (t) to the digital signal y (n). The inverse noise filter 120 is coupled to the analog-to-digital converter 110 for receiving the digital signal y (n).

The inverse noise filter 120 is used for processing the digital signal y (n) to generate a noise cancellation signal nc (n). In some embodiments, the noise cancellation signal nc (n) is used to reduce the effect of ambient noise (e.g., the noise signal v (t)). In some embodiments, the inverse noise filter 120 may be an adaptive filter.

The output circuit 130 includes an arithmetic circuit 132, a digital-to-analog converter 134, and an electroacoustic conversion device 136. The operation circuit 132 receives the noise cancellation signal nc (n), and mixes the noise cancellation signal nc (n) and the equalization signal a (n) to generate a mixing signal u (n). In some embodiments, the operation circuit 132 may be implemented by an adder and/or a synthesizer. The digital-to-analog converter 134 is coupled to the operation circuit 132 and is used for converting the mixing signal u (n). The electroacoustic conversion device 136 is coupled to the digital-to-analog converter 134, and is configured to output the mixing signal u (n) converted by the analog converter 134 as an audio output signal so (t). In some embodiments, the electroacoustic conversion device 136 may be implemented as a speaker.

In some embodiments, the equalizer circuit 140 is configured to adjust signal components of the input signal m (n) in each frequency band to generate different sound effects. In some embodiments, the equalizer circuit 140 is configured to adjust at least one parameter of the equalizer circuit 140 according to the equalized signal a (n), the mixing signal u (n), and the digital signal y (n) to determine the equalized signal a (n).

In the example of fig. 1, the equalizer circuit 140 includes an equalizer 141, an adaptive (adaptive) filter 142, an inverse noise filter 143, and an arithmetic circuit 144. The equalizer 141 is used for processing the input signal m (n) to generate an equalized output signal eo (n). In some embodiments, the input signal m (n) may be an audio signal provided by a music source.

The adaptive filter 142 is used for processing the equalized output signal eo (n) according to the mixing signal u (n), the digital signal y (n) and the proportional signal a (n). In some embodiments, adaptive filter 142 includes a filter circuit 142A and a control circuit 142B. The filter circuit 142A is configured to provide a transfer function sinv (z) to process the equalized output signal eo (n) to generate a filtered signal F1 (n). The control circuit 142B is used for adjusting at least one weight coefficient (e.g. w described later) according to the error signal e (n) (not shown) when the power of the equalization signal a (n) is greater than the predetermined value THD_k) To update the aforementioned transfer function sinv (z). The detailed operation will be described with reference to fig. 2A in the following paragraphs. In various embodiments, the adaptive filter 142 may be implemented by various digital circuits, such as a finite impulse response filter, etc., but the disclosure is not limited thereto.

In some embodiments, the inverse noise filter 143 is configured to process the equalized output signal eo (n) to generate a filtered signal F2 (n). The operation circuit 144 is coupled to the equalizer circuit 140 and the inverse noise filter 143 for receiving the filtered signal F1(n) and the filtered signal F2(n), respectively. In some embodiments, the operation circuit 144 is configured to subtract the filtered signal F2(n) from the filtered signal F1(n) to generate the aforementioned equalized signal a (n).

In some embodiments, the input signal m (n) is directly input to the operation circuit 132 without the equalizer circuit 140 to generate the mixing signal u (n). Under this condition, the audio processing apparatus 100 can be analyzed by using Z-transform to obtain the following formula (1):

where m (Z) is the Z-transform of the input signal m (n), y (Z) is the Z-transform of the digital signal y (n), h (Z) is the transfer function of the inverse noise filter 120, v (Z) is the Z-transform of the noise signal v (t) corresponding to the digital signal v (n), and s (Z) is the equivalent transfer function between the electroacoustic conversion device 136 and the acousto-electric conversion device 150. According to the above formula (1), the transfer function from the input signal m (n) to the adc 110 is s (z)/[1-s (z) h (z) ].

Therefore, in the embodiment with the equalizer circuit 140, the transfer function e (z) of the equalizer circuit 140 can be expressed as the following formula (2):

wherein d (z) is the transfer function of the equalizer 141, i.e. the target transfer function of the input signal m (n) outputted to the back of the human ear. In some embodiments, the transfer functions d (z) and h (z) are predetermined known values, and the transfer function s (z) varies with the position of the earphone or the shape of the ear. In some embodiments, the equalizer circuit 140 in fig. 1 can be designed according to the above equation (2).

In some techniques, the audio processing apparatus only adjusts the input signal m (n) by using an equalizer with a fixed transfer function. In these techniques, the equalizer is not dynamically adjusted according to the change of the transfer function s (z), so that the adjusted playing effect of the equalizer is reduced. In contrast to the above-mentioned techniques, by providing the adaptive filter 142, the transfer function s (z) can be estimated according to the mixing signal u (n) and/or the equalizing signal a (n), and the parameters of the adaptive filter 142 can be dynamically adjusted according to the estimated transfer function s (z). In this way, the audio processing apparatus 100 can dynamically adjust the final output sound effect according to the position of the earphone or different ear configurations.

Fig. 2A is a flow diagram of a method 200 as performed by the control circuit 142B of fig. 1, in accordance with some embodiments of the present disclosure.

In some embodiments, the control circuit 142B may be implemented by any hardware, software, or combination thereof for performing the method 200. In some embodiments, the hardware includes a processor, a microcontroller, an asic, or various types of digital signal processing circuits.

As shown in fig. 2A, in operation S210, the mixed signal u (n) is processed through an adaptive filter having a transfer function w (z) to estimate the transfer function S (z). In some embodiments, the adaptive filter may be implemented by a FIR filter, and the transfer function w (z) may represent w (z) ═ w₀+w₁ ^z-1+w₂ ^z-2+…+w_L-1z^-L+1。

In operation S220, the output of the adaptive filter with the transfer function w (z) is subtracted by the digital signal y (n) to generate an error signal e (n).

In operation S230, an adaptive algorithm is performed to update at least one weight coefficient w in the transfer function w (z) when the power of the equalized signal a (n) is greater than a predetermined value THD_k。

For example, referring to fig. 1, for an audio processing apparatus 100 that analyzes using Z-transform, the following equation (3) can be derived:

wherein A (Z) is the Z conversion of the equalized signals a (n). From the above equation, when the power of the equalization signal a (n) is larger than the power of the noise signal v (t), the following equation (4) is obtained:

under these conditions, the ratio of Y (z) to U (z) is S (z). In other words, when the power of the equalized signal a (n) is greater than the power of the digital signal V (n) corresponding to the noise signal V (t), the transfer function W (z) can be converged to the transfer function S (z). On the contrary, when the power of the equalized signal a (n) is smaller than the power of the digital signal V (n) corresponding to the noise signal V (t), the transfer function W (z) can only converge to 1/H (z).

Accordingly, based on the above relationship, it can be determined whether the power of the equalization signal a (n) is sufficiently higher than the power of the digital signal v (n) corresponding to the noise signal v (t) by setting the predetermined value THD. In some embodiments, the adaptive algorithm in operation 230 may be expressed as:

ifa[n]²＞THD:w_k＝w_k+μe(n)×U(n-k),k＝0～L-1；

else:w_k＝w_k

wherein the error signal e (n) is generated based on the mixing signal U (n) and the digital signal Y (n), e.g. the error signal e (n) is the difference between the output of the adaptive filter with the transfer function W (z) and the digital signal Y (n), w_kA [ n ] is a plurality of weight coefficients in the above transfer function W (z)]²Denoted as the power of the equalized signal a (n), and μ is the step-size.

With continued reference to fig. 2A, in operation S240, the fast fourier transform, the reciprocal operation and the inverse fast fourier transform are sequentially performed on the transfer function w (z) to adjust the transfer function sinv (z) of the filter circuit 142A, wherein the transfer function sinv (z) is associated with the reciprocal of the transfer function S (z). For example, in one embodiment, the transfer function Sinv (z) may be set substantially to the inverse of the transfer function S (z).

By operating the operations in S240, 1/S (z) (i.e., the transfer function sinv (z)) in the above equation (2) can be estimated. Equivalently, the equalizer circuit 140 can perform the operation of equation (2) to process the input signal m (n) to dynamically adjust the sound output effect according to the transfer function s (z).

Fig. 2B is a functional block diagram illustrating a method 200 for performing fig. 2A according to some embodiments of the present disclosure. In some embodiments, adaptive filter 142 may be implemented by multiple functional units of fig. 2B.

For example, the unit 210A may be configured to perform the operations S210, S220, and S230 of fig. 2A. Unit 220A may be configured to perform the aforementioned fast fourier transform operations (denoted as FFT) in operations S240 of fig. 2A. The unit 230A may be configured to perform the reciprocal operation (denoted as (1 /) in the operations S250 of fig. 2A. Unit 240A may be configured to perform the inverse fast fourier transform operation (denoted as IFFT) in operations S250 of fig. 2A. The functional block diagram of fig. 2B is merely an example, and the disclosure is not limited thereto.

Fig. 3 is a schematic diagram illustrating an audio processing device 300 according to some embodiments of the present disclosure. For ease of understanding, similar elements in fig. 1 and 3 will be designated with the same reference numerals.

Compared to fig. 1, the control circuit 142B of the audio processing apparatus 300 adjusts at least one weight coefficient w in the transfer function w (z) according to the equalized signals a (n) and the digital signals y (n) only_k. For example, compared to the control circuit 142B of fig. 1, the control circuit 142B in this example is configured to process the equalized signal a (n) through an adaptive filter having a transfer function w (z) in operation S210 to estimate the transfer function S (z). In this case, the error signal e (n) is generated based on the equalized signal a (n) and the digital signal y (n). For example, the error signal e (n) is the difference between the output of the adaptive filtering with the transfer function w (z) and the digital signal y (n). In other words, in this example, the aforementioned unit 210A in fig. 2B may estimate the transfer function s (z) without receiving the mixing signal u (n).

For example, according to the above equation (3), when the power of the equalized signal a (n) is greater than the power of the digital signal v (n) corresponding to the noise signal v (t), the following equation (4) can be obtained:

therefore, under these conditions, y (z)/a (z) ═ s (z)/[1-s (z) h (z) ]. Thus, the transfer function W (z) converges to S (z)/[1-S (z) H (z) ], and the transfer function Sinv (z) converges to 1/S (z) -H (z). Thus, when the transfer function 1/S (z) is sufficiently greater than the transfer function H (z), the transfer function Sinv (z) can be approximately equal to 1/S (z). Alternatively, when the transfer function 1/s (z) is smaller than the transfer function h (z), it can be analyzed that the error value of the transfer function sinv (z) is only about 6 dB. Accordingly, under any condition, the control circuit 142B in this example is still sufficient to estimate the transfer function sinv (z).

In the above embodiments, the circuits in the audio processing apparatus 100 and the audio processing apparatus 300 can be implemented by various types of digital signal processing circuits. Alternatively, various operations in the above embodiments may be set by a state machine and implemented via hardware, software, or a combination thereof. Various embodiments for implementing the components and functions of the

audio processing apparatus

100 and 300 are all within the scope of the present disclosure.

Fig. 4 is a flow diagram illustrating an audio processing method 400 according to some embodiments of the present disclosure. In operation S410, the digital signal y (n) is processed to generate a noise cancellation signal nc (n).

For example, as shown in fig. 1, the inverse noise filter 120 is configured to process a digital signal y (n) generated by converting an electronic signal e (t) through the adc 110 to generate a noise cancellation signal nc (n).

In operation S420, the mixing noise cancellation signal nc (n), the equalization signal a (n) generate a mixing signal u (n), and output an audio output signal so (t) based on the mixing signal u (n). For example, as previously described, operation S420 may be performed by the output circuit 130 of fig. 1.

In operation S430, at least one parameter is adjusted according to the equalization signal a (n) and the digital signal y (n) to determine the equalization signal a (n). For example, as previously described, the method 200 may be performed by the control circuit 142B of fig. 1 or 3 to adjust the at least one weight parameter w_kWherein the weight parameter w_kThe transfer function sinv (z) associated with filter circuit 142A. In this way, the equalization signal a (n) can be dynamically changed according to the position of the earphone or different ear configurations to maintain the final output sound effect.

The steps of the method 400 are exemplary only and not limited to the sequential execution of the above examples. Various operations under the method 400 may be added, substituted, omitted, or performed in a different order, as appropriate, without departing from the manner of operation and scope of various embodiments of the disclosure.

Although the present disclosure has been described with reference to the above embodiments, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the disclosure, and therefore, the scope of the disclosure should be determined by that defined in the appended claims.

Claims

1. An audio processing apparatus, comprising:

a first inverse noise filter for processing a digital signal to generate a noise cancellation signal;

an output circuit for mixing the noise cancellation signal and an equalization signal to generate a mixed signal and generating an audio output signal based on the mixed signal, wherein the digital signal is associated with the audio output signal and the equalization signal is generated by a first filtered signal and a second filtered signal; and

an equalizer circuit for adjusting at least one parameter of the equalizer circuit according to the equalized signal and the digital signal, wherein the equalizer circuit comprises:

an equalizer for processing an input signal to generate an equalized output signal; and

a plurality of filter circuits for processing the equalized output signal based on the at least one parameter to generate the first filtered signal from the equalized signal and the digital signal, and for processing the equalized output signal to generate the second filtered signal.

2. The audio processing apparatus according to claim 1, wherein the output circuit comprises:

an operational circuit for mixing the noise cancellation signal and the equalized signal to generate the mixing signal;

a digital-to-analog converter for converting the mixing signal; and

an electroacoustic conversion device for outputting the sound output signal according to the converted mixing signal.

3. The audio processing apparatus of claim 2, further comprising:

an acoustoelectric conversion device for generating an electronic signal based on a noise signal and the sound output signal; and

an analog-to-digital converter for converting the electronic signal to the digital signal.

4. The audio processing apparatus of claim 3 wherein a first transfer function exists between the electro-acoustic transducer and the acousto-electric transducer, the plurality of filter circuits comprises an adaptive filter having a second transfer function, the at least one parameter is at least one weight coefficient associated with the second transfer function, and the second transfer function is associated with being the inverse of the first transfer function.

5. The audio processing apparatus according to claim 1, wherein the plurality of filter circuits includes an adaptive filter, and the at least one parameter is at least one weight coefficient of the adaptive filter.

6. The audio processing apparatus according to claim 1, wherein the equalizer circuit comprises an arithmetic circuit for subtracting the second filtered signal from the first filtered signal to generate the equalized signal, and the plurality of filter circuits comprises:

an adaptive filter for processing the equalized output signal according to the mixing signal, the equalized signal and the digital signal to generate the first filtered signal; and

a second inverse noise filter for processing the equalized output signal to generate the second filtered signal.

7. The audio processing apparatus according to claim 6, wherein the adaptive filter comprises:

a filter circuit for providing a transfer function to process the equalized output signal to generate the first filtered signal; and

a control circuit for adjusting the at least one parameter according to an error signal when a power of the equalized signal is greater than a predetermined value to update the transfer function,

wherein the error signal is generated based on the mixing signal and the digital signal, and the at least one parameter is at least one weight coefficient associated with the transfer function.

8. The audio processing device of claim 1 wherein the equalizer circuit comprises an arithmetic circuit for subtracting the second filtered signal from the first filtered signal to generate an equalized signal, and

the plurality of filter circuits includes:

an adaptive filter for processing the equalized output signal according to the equalized signal and the digital signal to generate a first filtered signal; and

a second inverse noise filter for processing the equalized output signal to generate a second filtered signal.

9. The audio processing apparatus according to claim 8, wherein the adaptive filter comprises:

wherein the error signal is generated based on the equalized signal and the digital signal, and the at least one parameter is at least one weight coefficient associated with the transfer function.

10. An audio processing method, comprising:

processing a digital signal to generate a noise cancellation signal;

mixing the noise cancellation signal and an equalization signal to generate a mixed signal, and outputting an audio output signal based on the mixed signal, wherein the digital signal is associated with the audio output signal, and the equalization signal is generated by a first filtering signal and a second filtering signal;

adjusting at least one parameter according to the equalized signal and the digital signal;

processing an input signal to generate an equalized output signal; and

processing, by a plurality of filter circuits of an equalizer circuit, the equalized output signal based on the at least one parameter to generate the first filtered signal from the equalized signal and the digital signal, and processing the equalized output signal to generate the second filtered signal.