Introduction

The term Speaker Recognition consists of Speaker Identification which is the identification of the speaker speaking the current utterance and Speaker Verification that refers to the verification from the utterance of whether the speaker is who he claims to be (Almaadeed et al., 2015). Unlike to other biometric characteristics like fingerprints and faces, a human voice is a biometric characteristic that is not yet frequently employed for person identification. A system uses a recording of a speaker's speech to verify or ascertain the speaker's identification in automatic voice recognition, also referred to as speaker recognition (Mokgonyane et al., 2021).

There are two types of speaker recognition: text-dependent and text-independent. The process is known as text-dependent speaker verification when the lexicon of the spoken utterances is limited to a single word or phrase across all speakers, as opposed to text-independent speaker verification, where speakers are free to say whatever they want without having their utterance constrained.

In this paper the approach is based on text-dependent verification for the comparison of intonation pattern of speech signals. The basic idea is taken from the fact that the vocal tract of a speaker is what distinguishes the voice from each other. Each person's vocal tract is unique in terms of size and shape which creates differences in their pitch frequency, intensity and formant frequency (Chaubey et al., 2022).

A man produces sound waves whenever he speaks, when the vocal folds come together and vibrate due to passing of air through them during exhalation from the lungs, they produce sound. This vibration produces the sound wave. Sound waves are characterised on the basis of frequency, amplitude and wavelength. The frequency, also known as pitch, is the number of times per second that a sound pressure wave repeats itself. The (quasi-)periodic structure of voiced speech signals are approximated by the fundamental frequency of a speech signal, which is sometimes indicated by F0. The vocal folds, when appropriately tensed, create an oscillation in the airflow. The mean size of oscillations per second, measured in Hertz, is the fundamental frequency. Formant frequencies, also known as F1, F2, and F3, are created when this fundamental frequency is amplified or attenuated by different parts of the resonating body (Ali et al., 2006; Magdin et al., 2019).

Intensity is the energy of a sound wave degree of loudness associated with it. It is measured in decibels. The intensity helps us in knowing how a person talks certain words. The amplitude of the sound wave is nothing but its height or the maximum distance that a medium's vibrating particles are moved from their average position during sound production. Due to variations in amplitude and pitch in a speaker's voice over different recordings, no two digital signals are identical, even when the same words are spoken by the same speaker (Almaadeed et al., 2015).

The intonation patterns are evaluated in the software called Praat. Praat converts speech to digital signals. It consists of waveform view and a spectrogram view of a sound. The waveform view gives amplitude information over time whereas the spectrogram shows the frequency information over time, basically the amplitude is shown through the shadings. It enables us to determine a sound's pitch, intensity, and formant values. In this paper the examination of the intonation pattern of the words is done using these parameters. The findings and observations are listed in the table below, along with the conclusion (Bhrati and Bansal, 2025; Wirdayanthi, 2022).