Oljective: Write a Matlab program for speech de-noising. Add Additive White Gaus

Question

Oljective: Write a Matlab program for speech de-noising. Add Additive White Gaussian Ncise (AWGN) to clean speech signal. Analyze the plot of noisy speech by finding its Power Spectral Density (PSD). Apply Gaussian filter to noisy speech and examine t obained. How Gaussian filter de-noise the signal? Give brief explanation. Write a Matlab program to: (a) Read clean speech signal (b) Run and listen clean speech signal (c) Generate Additive White Gaussian Noise (d) Add White Gaussian Noise to speech signal (e) Run and Listen noisy speech signal (t) Plot clean speech signal and noisy speech signal using subplot command (g) Find Signal to Noise Ratio of noisy speech (h) Find Power Spectral Density of clean speech signal, noisy speech signal, and White Gaussian Noise Using Gaussian filter de-noise the noisy speech (Hint: You can also use any adaptive filter) (i) G) Run and Listen de-noised speech signal

Explanation / Answer

a) % to Read clean speech signal

file.clean='speech_data.wav'; %speech_data.wav is a wave file name of speech
[ clean.speech, fs] = audioread(file.clean);

b) %Run and listen clean speech signal

sound(clean.speech, fs)

c) & d)

y=awgn(clean.speech, 'linear'); % linear define powertype it can be in dB as well. this command will automatically generate and add additive white gaussian noise.

or

v= sqrt(0.1)*randn(1,length(clean.speech));

y=clean.speech+v;

e) %To run and listen noisy speech signal

sound (y,fs)

f) plot clean speech signal and noisy speech signal

N=length(y);

f=[0:N/2]*fs/N;

subplot(2,1,1);

plot(f,y); % plot noisy signal

subplot(2,1,2);

plot(f, clean.speech); %plot clean signal

g) find Signal to noise Ratio of noisy speech

snr_noisy_speech=snr(y);

Related Questions

Navigate

• Browse (All)
• Browse O
• Subjects

---

---

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.