preview

Essay On Preprocessing Techniques

Decent Essays
Open Document

2.1 Preprocessing Module In this module, the input signals is prepossessed first, here Hamming windowing technique is used and followed by the FFT [5]. Initially the input signal is spitted as overlapping frames, and each frame contains the duration of 0.025ms. The block diagram of preprocessing module is as shown in Fig1.

Preprocessing Module

Optimization Module

Spectral Filtering Module

Fig.1 Block diagram of speech signal enhancement
The input speech signal is denoted by S by having a total duration of T ms and the frames be represented by Fi, where 1 ≤ i ≤ T/0.025 each having 0.025 ms. It can be represented by S = {F1 F2……… Fn}, when n=T/0.025 the frames are windowed by using the hamming window technique. The hamming …show more content…

In PSO each member of the population is called particle, and each population is called a swarm.
PSO algorithm steps: Initially it generates a random population. In this case the initial population consists of value interval [0, 1]. Compute the position and velocity of each and every particle. Compute the best velocity for each particle and the best velocity for all particles in the iterations. Update the new velocity, add it to the swarm particle and get the new particle.
Vt+1=vt+1/2αvt-1+1/6α(1-α)vt-2+1/24α(1-α)(- α)vt-3+Ψ_1 (ρ_b-ρ)+Ψ_2 (ω_b-ρ)…..(3) ρ_(t+1)=ρ_t+v_(t+1) ………….. (4) After updating all the particles, evaluate using fitness function is satisfied, the process ends otherwise the whole process is repeated from step3.
The fitness [1] in this paper depends on three terms. For calculating the fitness in this case, the values are converted to zero or one. It can be represented by z, if z > 0.5 it is converted to 1, otherwise 0. The initial noise power spectrum is denoted by Λ and noise spectrum variance is denoted by spectrum distance can be calculated using equation 5.
〖SD〗^((t))=20 log_10⁡〖ᴧ^t 〗- log‖W_j^((i)) ‖,where 0< m < M-1. After the windowing technique followed by the Fast Fourier transforms (FFT) frequency domain signal is achieved. Let the input windowed signals in the ith frame be represented as w_0^((i)),w_1^((i)),…….,w_(M-1)^((i))and Fourier transform is given by: w_k^((i) )= ∑_(k=0)^(M-1)▒w_n^((i) ) e^(-i2πk

Get Access
Related
Get Access