ANALYSIS OF SPEECH SIGNAL

ABSTRACT







V. KRISHNA PRIYA (13004101)
B. VAMSI KRISHNA (13004109)
N. SAI RANADHEER GUPTA (13004115)

1
ANALYSIS OF SPEECH SIGNAL
1. SPEECH: Speech sound is a sensation of air pressure
vibrations produced by air exhaled from the lungs and modulated
and shaped by the vibrations of the glottal cords and the resonance
of the vocal tract as the air is pushed out through the lips and nose.
Speech is an immensely information-rich signal exploiting
frequency-modulated, amplitude-modulated and time-modulated
carriers. The speech energy above 4 kHz mostly conveys audio
quality and sensation. Just as the written form of a language is a
sequence of elementary alphabet, speech is also a sequence of
elementary acoustic sounds or symbols known as phonemes that
convey the spoken form of a language.
The information conveyed in speech includes the followings:
(a) Acousticphonetic symbols,
(b) Prosody,
(c) Gender information,
(d) Age,
(e) Accent,
(f) Speakers identity and
(g) Emotion and health.
Human speech production system consists of the lungs, larynx, vocal tract
cavity, nasal cavity, teeth, lips, and the connecting tubes.
There are two broad types of
Speech sounds as shown in Figure: voiced sounds like an e pronounced
as iy, and unvoiced sounds like s.



1





Glottal Pulse Model for Voiced Signals
Figure shows the Liljencrants-Fant (LF) model of a glottal pulse and its derivative.

MatLab Program:
function GlottalLF()

The interval of each sound wave has different frequency in its sub-
sections. Our paper has made an analysis of two Matlab functions
namely GenerateSpectrogram.m and MatrixToSound.m, in order to
analyze and synthesis the speech signals. The first Matlab code section
GenerateSpectrogram.m record the user input sound for user (more
precisely from the source) defined duration and asks required parameters
for computation of spectrogram and returns a matrix with frequency as
rows and time as column and corresponding matrix element as amplitude
of that frequency. MatrixToSound.m uses the method of additive
synthesis of sound to generate sound from the user defined matrix with
frequencies as its rows and time as its columns. Sound recording is an
electrical or mechanical inscription of sound waves, such as spoken
voice, singing, instrumental music, or sound effects. The two main
classes of sound recording technology are analog recording and digital

2
recording. Acoustic analog recording is achieved by a small microphone
diaphragm that can detect changes in atmospheric pressure (acoustic
sound waves) and record them as a graphic representation of the sound