Audio signal single echo effect processing

bqgup

Audio signal single echo effect processing [Copy link]

<i class="pstatus">This post was last edited by bqgup on 2020-8-4 18:05</i> # Audio signal single echo effect processing## 1. Echo signal principle#### When a sound signal encounters an obstacle during propagation, part of the sound energy will be absorbed, and the other part will be reflected back and can be heard by the human ear. This repetition of sound caused by the reflection of sound waves is called echo. A single echo is formed by a single reflection of the original sound, and multiple echoes are formed by the superposition of multiple single echoes. Sometimes we shout into the valley, and there is a response from the other side of the valley. This is the echo phenomenon. In reality, sound will lose energy during propagation, so the echo sound we hear each time is weakening, and there is a delay. These are the general principles of echo. ## 2. Echo algorithm design### 2.1 Audio signal#### Get the audio file mentioned in the previous post on one end. The link of the previous post->[Exploration of the difference between left and right channel signals of audio](https://en.eeworld.com/bbs/thread-1133721-1-1.html "Exploration of the difference between left and right channel signals of audio")<- We use the audio obtained in this post and upload the audio file again: 
<ignore_js_op>

<a href="javascript:;" class="loginf">段坤我吃定了截取.wav</a>

<em class="xg1">(1.6 MB, downloads: 19)</em>
</span>
<div class="tip tip_4" id="attach_492823_menu" style="position: absolute; display: none" disautofocus="true">
<div class="tip_c xs0">
<div class="y">2020-8-4 17:21 上传</div>
点击文件名下载附件

</div>
<div class="tip_horn"></div>
</div>

</ignore_js_op>
 ### 2.2 Algorithm design#### Suppose the original audio signal is a sampled sequence x(n), the sampling frequency is Fs, the delay time is tao, y(n) represents a single echo signal sequence, and the echo intensity coefficient is k, then y(n)=x(n)+kx(n-tao*Fs) #### According to the above echo principle, the matlab program is designed as follows: ``` %%%% Echo effect processing%%%% function ECHO_Pro() [y,Fs] = audioread('E:\8、matlab\echo voice changer\Duan Kun I am determined to intercept.wav'); y_left = y(:,1); %% left channel signal tao = 0.6; %% delay time k = 0.55; %% echo intensity coefficient Ndelay = fix(tao*Fs); %% delay sequence length for m = 1 : 1 : length(y)+Ndelay; %% echo signal processing if(m <= Ndelay) y_delay = 0; else y_delay(m) = k*y_left(m-Ndelay); %% the received sound will be attenuated to a certain extent, and noise will be introduced end end for n = 1 : 1 : length(y)+Ndelay; %% original sequence expansion if(n <= length(y)) y_left_k(n) = y_left(n); else y_left_k(n) = 0; end end y_echo = y_left_k + y_delay; sound(y_echo,Fs) %% Play audio with echo effect``` #### Run the above code to feel the beauty of the echo effect. #### Let's take a look at the time domain and frequency domain changes of the echo audio. The program has added several new drawing functions: ``` %%%% Echo effect processing + time domain and frequency domain display%%%% function ECHO_Pro() [y,Fs] = audioread('E:\8、matlab\echo voice change processing\Duan Kun I am determined to intercept.wav'); y_left = y(:,1); %% Left channel signal tao = 0.6; %% Delay time k = 0.55; %% Echo intensity coefficient Ndelay = fix(tao*Fs); %% Delay sequence length for m = 1: 1: length(y)+Ndelay; %% Echo signal processingif(m <= Ndelay) y_delay = 0; else y_delay(m) = k*y_left(m-Ndelay); %% The received sound will be attenuated to a certain extent, and noise will be introducedend end for n = 1 : 1 : length(y)+Ndelay; %% Original sequence expansionif(n <= length(y)) y_left_k(n) = y_left(n); else y_left_k(n) = 0; end end y_echo = y_left_k + y_delay; sound(y_echo,Fs) %% Play audio with echo effectLength = length(y_echo); %% Signal lengthyy = abs(fft(y_echo))./Length; f=(0:Length-1)*Fs./Length; %% Actual frequency transformationsubplot(311); plot(f(1:Length/2),yy(1:Length./2)); title("N-DFT transform amplitude-frequency response unilateral"); xlabel("f/Hz"); grid on f=f-Fs./2; % Shift subplot(312); plot(f,fftshift(yy)); title("N-DFT transform amplitude-frequency response bilateral") xlabel("f/Hz") subplot(313); m = 1 : 1 : (Length); plot((m)/Fs,y_echo); title('Echo time domain change'); xlabel('t'); ylabel('y_echo') ``` #### The results are as follows: 
<ignore_js_op>

<div class="tip tip_4 aimg_tip" id="aimg_492824_menu" style="position: absolute; display: none" disautofocus="true">
<div class="xs0">
<p><strong>111.jpg</strong> <em class="xg1">(37.97 KB, downloads: 0)</em></p>
<p>
<a href="javascript:;" class="loginf" >download attach</a>

<a href="javascript:;" onclick="showWindow(this.id, this.getAttribute('url'), 'get', 0);" id="savephoto_492824" url="home.php?mod=spacecp&ac=album&op=saveforumphoto&aid=492824&handlekey=savephoto_492824">save to album</a>

</p>

</div>
<div class="tip_horn"></div>
</div>

</ignore_js_op>