Solving problem of the Virtual bass enhancement (VBE) of audio
7 次查看(过去 30 天)
显示 更早的评论
Hello recently I am trying to implement VBE for the enhancement of audio signal, I used the implementation of Non linear device(NLD) and Phase Vocoder(PV) . However the output seemS to be little bit distorted. Any one knows one of the best way to implement an NLD method with non distortion, or if he knows the implementation of maxxbass, Rbass or bass boost ? Please I need your help, thank you. Below is the matlab file for my NLD method. However the output sound tends to have some distortion like timbre has little bit changed. Anyone knows the best ways to implement a NLD?
%My VBE-NLD
clear all;
%close all
clc;
%%Load audio sample
%Fs=44100;
%audio=audioread('hiphop.wav');
%audio=audioread('sample.mp3');
%audio=audioread('audios/jazz.wav');
%audio=audioread('audios/classical.wav');
%audio=audioread('audios/hiphop.wav');
%audio=audioread('audios/rock.wav');
%[audio, Fs]=audioread('audios/drum.wav');
%[audio, Fs]=audioread('audios/8_micarray.wav');
[audio, Fs]=audioread('audios/song1.mp3');
%audio=audioread('audios/rock.wav');
%audio = sum(audio, 2) / size(audio, 2); %stereo to mono
%%%parameters
Fs=Fs;
x=audio;
Fcc=150; % cut-off frequency (Hz)
GdB=3; % NLD gain for transient calibration
%HPF
Fc=150; %cut-off frequency
N11=5;
[b,a]= High_pass(N11,Fc,Fs);
yt_high=Myiir_filter(b, a, x);
%LPF
Fc=150;
N11=5;
[b,a]= Low_pass(N11,Fc,Fs);
yt_low=Myiir_filter(b, a, x);
%y2=yt_high+yt_low;
%Apply gain
%G=10^(GdB/20);
%y3=G*yt_low;
%y4=yt_high+y3;
%y5=HPF_final(y4,Fc, Fs);
%NLD
H_hwr=[0.0278 0.5 1.8042 0 -6.0133 0 12.2658 0 -11.8891 0 4.3149];
H_fexp1=[0 1.4847 0 -1.4449 0 2.4713 0 -2.4234 0 0.9158 0];
h=H_hwr;
for n=1:11
if n==1
yt_low_proc=h(n);
else
yt_low_proc=yt_low_proc+h(n)*(yt_low.^(n-1));
end
end
h=H_fexp1;
for n=1:11
if n==1
yt_low_proc_2=h(n);
else
yt_low_proc_2=yt_low_proc_2+h(n)*(yt_low_proc.^(n-1));
end
end
%yt_low_proc_1=atsr(yt_low);
%yt_low_proc_2=abs(yt_low_proc_1);
%yt_low_proc_2=asqrt(yt_low);
%yt_low_proc_2=abs(yt_low_proc_1);
%yt_low_proc_2=my_full_wav_rect(yt_low);
%yt_low_proc_2=my_half_wav_rect(yt_low);
%my_softclipping(x, k)
%BPF
NBPF1 = 200; % Order
NBPF2 = 100; % Order
Fstop = 2*Fcc/4; % Stopband Frequency
Fpass = Fcc+ Fcc/4; % Passband Frequency
Fpass2=2*Fcc;
Fstop2=5*Fcc;
Wstop = 1; % Stopband Weight
Wstop2 = 0.001; % Stopband Weight
Wpass = 1; % Passband Weight
Wpass2 = 1; % Passband Weight
dens = 20; % Density Factor
b = firpm(NBPF1, [0 Fstop Fpass Fsreduced/2]/(Fsreduced/2), [0 0 1 1], [Wstop Wpass], ...
{dens});
Hd = dfilt.dffir(b);
yt_low_proc_bpf=filter(Hd,yt_low_proc_2);
b = firpm(NBPF2, [0 Fpass2 Fstop2 Fsreduced/2]/(Fsreduced/2), [1 1 0 0], [Wpass Wstop], ...
{dens});
Hd = dfilt.dffir(b);
yt_low_proc_bpf=filter(Hd,yt_low_proc_bpf);
%delay adjustment
delay_low=round((NBPF1+NBPF2)/2);
yt_high=[zeros(delay_low,1); transpose(yt_high)];
if length(yt_high)>length(yt_low_proc_bpf)
yt_high=yt_high(1:length(yt_low_proc_bpf));
else
yt_low_proc_bpf=yt_low_proc_bpf(1:length(yt_high));
end
%Apply gain
G=10^(GdB/20);
yt_low_proc_gain=G*yt_low_proc_bpf;
%Reconstruction
yt_proc=yt_high+transpose(yt_low_proc_gain);
%yt_proc=HPF_final(yt_proc, Fcc,Fs);
xfilt=MyHPF_final(x,Fcc,Fs);
yfilt=MyHPF_final(yt_proc,Fcc,Fs);
y_filt_low=MyLPF_final(x,Fcc,Fs);
0 个评论
回答(0 个)
另请参阅
类别
在 Help Center 和 File Exchange 中查找有关 Spectral Measurements 的更多信息
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!