CN102760461A - Volume adaptive control audio playing equipment and method - Google Patents

Abstract

The invention discloses a volume adaptive control audio playing equipment and method. The delay of a back-feed sound is estimated through the cross-correlation algorithm, the delay of a decoding audio signal is taken as a reference input for adaptive processing, and a sound signal detected by a noise detecting module is taken as a signal input for adaptive processing. The adaptive processing adopts an FIR (finite impulse response) filter and the NLMS (normalized least mean squares) adaptive filtering algorithm. Through adaptive processing, the normally played back-feed sound is inhibited, the real noise signal can be extracted from a noise sensor, and the playing volume of the playing equipment can be regulated after performing weighted moving average for the noise level. The volume adaptive control audio playing equipment and method disclosed by the invention are suitable for occasions with violent environmental noise variation.

Description

A kind of self-adaptation is regulated the audio-frequence player device and the method for volume

Technical field

The present invention relates to suppress in a kind of audio-frequence player device the environment noise test technology and the adaptive gain regulation technology of broadcast sound; Particularly relate to a kind of back coupling that can suppress or eliminate the broadcast sound in the noise transducer, regulate the audio-frequence player device and the method for volume according to the neighbourhood noise size adaptation.

Background technology

Change more violent occasion playing audio signal in neighbourhood noise, phenomenon such as the energy that neighbourhood noise and normal play sound can occur does not match, and causes audience's auditory fatigue, and the sense of hearing is unintelligible.If neighbourhood noise is excessive, signal to noise ratio (S/N ratio) is too low, and the audience need pick up one's ears and just can not hear, and causes auditory fatigue; If neighbourhood noise is lower, excessive play sound and can damage ear hearing.

One of gordian technique that realizes Gain Automatic control is the detection technique of environmental noise power.Sound wave is propagated in air, not only comprises neighbourhood noise in the walkaway sensor, also comprises the normal play sound signal, has only the echo with the normal play audio frequency in the sensor to curb, and just can obtain real environmental noise power.

Existing a kind of method avoids picking up the broadcast sound through the special mounting of ambient noise sensor, and the noise energy of calculating noise sensor gains with this parameter controls playing device then.In the practical application, be difficult to satisfy this requirement, sometimes in addition the broadcast sound that picks up of noise transducer all can be better than neighbourhood noise.Move back and ask time, the another kind of application is to play intermittently, promptly play when suspending, detection noise, and in proper order noise level as the gain control foundation.This method is not suitable for continuous uninterrupted again and plays occasion.

Perfected process is that signal Processing is carried out in output to noise transducer, suppresses the broadcast sound in the noise transducer pickup, and this not only need design suitable algorithm, also needs hardware calculate platforms such as DSP or flush bonding processor.There are some patents to relate to the narration of this respect at present, but do not have public algorithm and implementation method.

Summary of the invention

The present invention solves the technical matters that existing in prior technology is difficult to accurately obtain the environmental noise size, provides a kind of and can filtering plays sound, accurately obtains audio-frequence player device and the method that the self-adaptation of environmental noise size is regulated volume.

The present invention is directed to above-mentioned technical matters mainly is able to solve through following technical proposals: a kind of self-adaptation is regulated the audio-frequence player device of volume, comprising:

Processor module is used for entire equipment is controlled;

Noise detection module is used for the testing environment noise and sends to processor module;

Power amplifier module is used for the play signal of processor module output is amplified;

Loudspeaker are used for the play signal after the power amplifier module amplification is play;

The COBBAIF module; Said COBBAIF module is connection processing device module, power amplifier module and noise detection module simultaneously; The detected signal of noise detection module is passed to processor module, and the play signal with processor module passes to power amplifier module simultaneously.

As preferably, the audio-frequence player device that self-adaptation is regulated volume also comprises wireless network module, and said wireless network module is connected with said processor module.

As preferably, said processor module comprises USB interface and SD card.

As preferably, the audio-frequence player device that self-adaptation is regulated volume also comprises the analogue audio frequency module, and said analogue audio frequency module is connected with processor module through the COBBAIF module.

The source of sound of playing sound can be the audio file that is stored in through in USB interface or SD card and the memory device that processor module is connected; Also can be the network audio Streaming Media through the wireless network module input, can also be the local analogue audio frequency through the input of analogue audio frequency module.

A kind of self-adaptation is regulated the audio frequency playing method of volume, may further comprise the steps:

Step 1, play audio file from source of sound;

Step 2, testing environment noise, the size of judgement neighbourhood noise;

Step 3, play the volume of sound according to the neighbourhood noise size adjustment.

As preferably, in the said step 2, the testing environment noise is specially:

A, estimate to feedback the time-delay of sound through cross correlation algorithm;

B, to reference to the input pre-service of delaying time;

C, suppress the back coupling sound in the detected voice signal of noise transducer with the NLMS adaptive filter algorithm, thereby extract noise signal, the later reference-input signal of the pre-service that is about to delay time is through the FIR sef-adapting filter;

Reference-input signal is the sound signal of decoding output from the audio file of storages such as SD card, USB flash disk, or passes through the sound signal of network reception and decoding output.

As preferably, in the said step 3, the volume of playing sound according to the neighbourhood noise size adjustment is specially: at first noise level is carried out the weighting running mean and handle, then according to auditory properties, gain is regulated by the arc tangent curve with noise level.

As preferably, said weighting relation is:

${w^{\′}}_{\mathrm{noise}}\left(n\right)=\frac{1}{35}[-3w_{\mathrm{noise}}(n-2)+12w_{\mathrm{noise}}(n-1)+17w_{\mathrm{noise}}\left(n\right)+12w_{\mathrm{noise}}(n+1)-3w_{\mathrm{noise}}(n+2)]$

w _NoiseBe the noise level that calculates according to ambient noise signal η (n), w ' _NoiseIt is the later noise level value of weighting.

As preferably, said gain-adjusted curve is:

A＝f(w′ _noise)＝tan ^-1(w′ _noise)

A is a system-gain.

The present invention adopts auto-adaptive filtering technique to suppress the broadcast sound in the noise transducer, and the sound signal of player inner decoding output is imported as a reference, and the input of noise transducer is imported as signal.Adaptive echo elimination algorithm commonly used has least-mean-square error algorithm LMS, normalization minimum mean-square ERROR ALGORITHM NLMS, RLS RLS etc.The characteristics of LMS adaptive algorithm are that complexity is low, and robustness is good, and steady-state error and speed of convergence can not be optimum simultaneously.NLMS is a kind of important variable step LMS algorithm, and it is being inherited on the LMS algorithm advantage basis, and when error was big, step-length was bigger, can accelerate speed of convergence; When error was stablized, step-length was less, can reduce steady-state error.The RLS speed of convergence is faster than LMS, and steady-state error is also less, but computation complexity is more a lot of greatly than LMS and NLMS algorithm.

Sef-adapting filter adopts FIR type digital filter in the NLMS algorithm, and filter order is a key parameter, it and the residing sound field environmental correclation of player, its size can be from tens rank extremely thousands of rank do not wait.Wave-wave device order is on the low sidely fixed, suppresses effect and reduces; If order is too highly fixed, can increase computation complexity in vain again.Because the sound field environment of practical application has very big uncertainty, bring difficulty for the setting of sef-adapting filter order.

In order to reduce computation complexity; The present invention at first estimates to play the time-delay of sound through cross correlation algorithm, to decoding output audio signal (promptly with reference to the input) pre-service of delaying time, so just can adopt 128 fixing rank FIR sef-adapting filters then; Reduce computational complexity; Extract noise signal with the NLMS adaptive filter algorithm more at last, increased speed of convergence, reduced steady-state error.

After obtaining the noise level estimation, must the suitable gain-adjusted control algolithm of design.Gain-adjusted algorithm and automatic gain control (AGC) algorithm are similar, must and regulate between the stationarity compromise in response speed.The present invention at first carries out the weighting running mean to noise level and handles, and reduces the violent degree of change in gain, and then according to auditory properties, gain is regulated by the arc tangent curve with noise level.

The present invention can be applicable to occasions such as outdoor DAB broadcast such as bus station, railway station, subway platform, bus-stop; And occasions such as indoor digital voice playing such as bar, coffee shop, background music; Use the present invention and can obtain constant signal to noise ratio (S/N ratio), reach best audio frequency sense of hearing and audio frequency definition.

The substantial effect that the present invention brings is that (1) proposes a kind of integration algorithm that ambient noise level is regulated the player gain automatically of following, and satisfies various broadcast level control needs with the self-reacting occasion demand of neighbourhood noise.(2) extract the time-delay that noise level algorithm use cross correlation algorithm is estimated audio plays, and then adopt the NLMS adaptive filter algorithm to eliminate the back coupling of audio plays; (3) adopt weighting running mean algorithm to reduce the undulatory property of gain-adjusted, adopt arctan function to regulate the relation between gain and the noise level.

Description of drawings

Fig. 1 is a kind of structural representation of the present invention;

Fig. 2 is a kind of algorithm block diagram of the present invention;

Fig. 3 is the block scheme of a kind of auto-adaptive fir filter of the present invention;

Fig. 4 is a kind of automatic gain control flow chart of the present invention;

Fig. 5 is the graph of a relation between a kind of noise level of the present invention and the system-gain;

Among the figure: 101, processor module, 102, SD card or USB flash disk, 103, wireless network module, 104, the analogue audio frequency module, 105, the COBBAIF module, 106, power amplifier module, 107, loudspeaker, 108, user interface, 109, noise detection module.

Embodiment

Pass through embodiment below, and combine accompanying drawing, do further bright specifically technical scheme of the present invention.

Embodiment: a kind of self-adaptation of present embodiment is regulated the audio-frequence player device of volume; As shown in Figure 1, comprise processor module 101, COBBAIF module 105, power amplifier module 106, loudspeaker 107, noise detection module 109, analogue audio frequency module 104 and wireless network module 103.COBBAIF module 105 is connection processing device module 101, power amplifier module 106, noise detection module 109 and analogue audio frequency module 104 simultaneously; Noise detection module 109 detected signals are passed to processor module 101, and the play signal with processor module 101 passes to power amplifier module 106 simultaneously.Power amplifier module 106 is play through loudspeaker 107 after play signal is amplified.Processor module 101 includes USB interface and SD card.The source of sound of playing sound can be to be stored in the USB flash disk that is connected with processor module through USB interface or SD card or the audio file in the SD card 102; Also can be the network audio Streaming Media through wireless network module 103 inputs, can also be the local analogue audio frequency through 104 inputs of analogue audio frequency module.

Processor module 101 also is connected with user interface 108, and the user can carry out the parameter setting through 108 pairs of equipment of user interface.

Noise detection module 109 detected sound signals also comprise the back coupling of playing sound except that the environment noise.

A kind of self-adaptation of present embodiment is regulated the audio frequency playing method of volume, may further comprise the steps:

Step 1, play audio file from source of sound;

Step 2, testing environment noise, the size of judgement neighbourhood noise;

Step 3, play the volume of sound according to the neighbourhood noise size adjustment.

The testing environment noise is specially:

A, estimate to feedback the time-delay of sound through cross correlation algorithm;

B, to decoded audio signal (with reference to the input) pre-service of delaying time;

C, suppress the back coupling sound in the detected voice signal of noise transducer, thereby extract noise signal with the NLMS adaptive filter algorithm.

The volume of playing sound according to the neighbourhood noise size adjustment is specially: at first noise level is carried out the weighting running mean and handle, then according to auditory properties, gain is regulated by the arc tangent curve with noise level.

Back coupling or echo that present embodiment utilizes cross correlation algorithm and adaptive filter algorithm to suppress to play sound.X (n) is the normal play sound signal, and f (n) is the detected sound signal of noise detection module, comprises ambient noise signal η (n) in the f (n) and plays echoed signal y (n)=T [x (n)] that sound produces through the space, promptly

f(n)＝y(n)+η(n)

As shown in Figure 2, the simple crosscorrelation r of f (n) and x (n) calculates in elder generation _Xf(n, n+ τ), because x (n) and y (n)=T [x (n)] have similarity, the peak point of cross correlation function promptly is to feedback to delay time, and analyzes cross correlation results, obtains the echo time-delay K of x (n).X (n) is through K sampled point x of time-delay _D(n)=x (n-K).x _D(n) get into sef-adapting filter with f (n) and carry out the echo inhibition, obtain ambient noise signal η (n).

Fig. 3 is the implementation structure of auto-adaptive fir filter.The FIR wave filter adopts transversal type structure, and its unit impulse response H (n) is:

H(n)＝{h _n1，h _n2,..,h _nM}

X _D(n) be through the vector of the input time delay after the cross correlation process:

X _D(n)＝{x _D(n),x _D(n-1),...,x _D(n-M+1)}

is that the echo of audio plays is estimated:

$\hat{y}\left(n\right)=\underset{i=1}{\overset{M}{\mathrm{\Σ}}}{h}_i\left(n\right)\·x_D(n-i+1)=H{\left(n\right)}^T\·X_D\left(n\right)$

The objective function that suppresses the echo algorithm is:

J＝min{||η(n)|| ²}， $\mathrm{\η}\left(n\right)=f\left(n\right)-\hat{y}\left(n\right)$

The present invention adopts variable step NLMS algorithm to make objective optimization, and the recursion formula of sef-adapting filter unit impulse response H (n) is:

$H(n+1)=H\left(n\right)+\mathrm{\μ}\·\frac{\∂{\left|\right|\mathrm{\η}\left(n\right)\left|\right|}^2}{\∂H}=H\left(n\right)-2\mathrm{\μ}\left(n\right)\·\mathrm{\η}\left(n\right)\·X_D\left(n\right)$

$\mathrm{\&mu;}(n+1)=\left\{\begin{array}{c}{\mathrm{\&mu;}}_{\min },\mathrm{\&mu;}<{\mathrm{\&mu;}}_{\min }\\ \mathrm{\&alpha;}\&CenterDot;\mathrm{\&mu;}\left(n\right)+\mathrm{\&beta;}\&CenterDot;{\left|r_{\mathrm{\&eta;x}}\left(n\right)\right|}^2\\ {\mathrm{\&mu;}}_{\max },\mathrm{\&mu;}>{\mathrm{\&mu;}}_{\max }\end{array}\right.$

Wherein, α is the step-length gene, and between 0.95 ~ 0.999, the big more steady-state error of this value is big more usually for interval.β is the instantaneous energy of noise and sound signal cross correlation function.When noise and sound signal were divided deviation, the simple crosscorrelation instantaneous energy was big, and is big to the step-length contribution, can accelerate speed of convergence.The β interval approaches 0 usually between 0 ~ 1, be typically value 10 ^-4~ 10 ^-6, convergence stability is inversely proportional to this value, and speed of convergence is directly proportional with this value.

Fig. 4 is an automatic gain control AGC flow process, w _NoiseBe the noise level that calculates according to ambient noise signal η (n) (unit: decibel), noise level is carried out 5 weighting running means, the weighting relation is:

${w^{\&prime;}}_{\mathrm{noise}}\left(n\right)=\frac{1}{35}[-3w_{\mathrm{noise}}(n-2)+12w_{\mathrm{noise}}(n-1)+17w_{\mathrm{noise}}\left(n\right)+12w_{\mathrm{noise}}(n+1)-3w_{\mathrm{noise}}(n+2)]$

Through transfer function, the noise gain after level and smooth is converted to the yield value of audio plays.

Fig. 5 is the relation curve between noise level and the system-gain.W ' _NoiseBe the later noise level value of weighting running mean, A is a system-gain, is the gain-adjusted curve,

A＝f(w′ _noise)＝tan ^-1(w′ _noise)

It is reference noise level and reference gain value that coordinate (0,0) is located.Usually deciding the reference noise level is 60db (open air), and the reference gain value is 0db.The gain-adjusted scope is to allow system-gain A on the basis of reference point, to increase 12dB, or reduces 12dB.The dynamic adjustments scope is 24dB.

Present embodiment is used the control system realization said process that flush bonding processor constitutes.

Specific embodiment described herein only is that the present invention's spirit is illustrated.Person of ordinary skill in the field of the present invention can make various modifications or replenishes or adopt similar mode to substitute described specific embodiment, but can't depart from spirit of the present invention or surmount the defined scope of appended claims.

Although this paper has used terms such as walkaway, inhibition echo morely, do not get rid of the possibility of using other term.Using these terms only is in order to describe and explain essence of the present invention more easily; It all is contrary with spirit of the present invention being construed to any additional restriction to them.

Claims (9)

1. the audio-frequence player device of a self-adaptation adjusting volume is characterized in that, comprises

Processor module is used for entire equipment is controlled;

Noise detection module is used for the testing environment noise and sends to processor module;

Power amplifier module is used for the play signal of processor module output is amplified;

Loudspeaker are used for the play signal after the power amplifier module amplification is play;

The COBBAIF module; Said COBBAIF module is connection processing device module, power amplifier module and noise detection module simultaneously; The detected signal of noise detection module is passed to processor module, and the play signal with processor module passes to power amplifier module simultaneously.

2. self-adaptation according to claim 1 is regulated the audio-frequence player device of volume, it is characterized in that also comprise wireless network module, said wireless network module is connected with said processor module.

3. self-adaptation according to claim 1 and 2 is regulated the audio-frequence player device of volume, it is characterized in that said processor module comprises USB interface and SD card.

4. self-adaptation according to claim 1 and 2 is regulated the audio-frequence player device of volume, it is characterized in that, also comprises the analogue audio frequency module, and said analogue audio frequency module is connected with processor module through the COBBAIF module.

5. a self-adaptation is regulated the audio frequency playing method of volume, it is characterized in that may further comprise the steps:

Step 1, play audio file from source of sound;

Step 2, testing environment noise, the size of judgement neighbourhood noise;

Step 3, play the volume of sound according to the neighbourhood noise size adjustment.

6. want 5 described a kind of self-adaptations to regulate the audio frequency playing method of volume according to right, it is characterized in that in the said step 2, the testing environment noise is specially:

A, estimate to feedback the time-delay of sound through cross correlation algorithm;

B, to reference to the input pre-service of delaying time;

C, suppress the back coupling sound in the detected voice signal of noise transducer, thereby extract noise signal with the NLMS adaptive filter algorithm.

7. regulate the audio frequency playing method of volumes according to claim 5 or 6 described a kind of self-adaptations; It is characterized in that; In the said step 3; The volume of playing sound according to the neighbourhood noise size adjustment is specially: at first noise level is carried out the weighting running mean and handle, then according to auditory properties, gain is regulated by the arc tangent curve with noise level.

8. a kind of self-adaptation according to claim 7 is regulated the audio frequency playing method of volume, it is characterized in that, said weighting relation is:

${w^{\&prime;}}_{\mathrm{noise}}\left(n\right)=\frac{1}{35}[-3w_{\mathrm{noise}}(n-2)+12w_{\mathrm{noise}}(n-1)+17w_{\mathrm{noise}}\left(n\right)+12w_{\mathrm{noise}}(n+1)-3w_{\mathrm{noise}}(n+2)]$

w _NoiseBe the noise level that calculates according to ambient noise signal η (n), w ' _NoiseIt is the later noise level value of weighting.

9. a kind of self-adaptation according to claim 7 is regulated the audio frequency playing method of volume, it is characterized in that said gain-adjusted curve is:

A＝f(w′ _noise)＝tan ^-1(w′ _noise)

A is a system-gain.

Images