International Journal on Future Revolution in Computer Science & Communication Engineering ISSN: 2454-4248 Volume: 4 Issue: 2 461 – 464 _______________________________________________________________________________________________ 461 IJFRCSCE | February 2018, Available @ http://www.ijfrcsce.org _______________________________________________________________________________________ Assessment of Respiratory Disorders Using Speech Parameters Poonam Shrivastava 1 , Neeta Tripathi 2 ,Bikesh Kumar Singh 3 1 Department of Electronics and Telecommunication, SSTC(SSGI), Bhilai, India 2 Department of Electronics and Telecommunication, SSTC (SSEC), Bhilai, India 3 Bio Medical Department, NIT, Raipur, India Abstract:In today‟s scenario detection of diseases using voice analysis has become one of the important research topic. In order to determine if the person is suffering from some kind of illness, the information extracted by voice analysis about the person‟s health c an be helpful, as a number of pathological disorders are related with nasal, neural, respiratory and larynx diseases etc. In this paper, the analysis of voice parameters is done in order to differentiate between normal and affected person. The speech samples are recorded and Pitch , formant frequencies, intensity was extracted from the speech signals. The mean values variation of these parameters is used in identifying the difference between normal and affected person. Keywords— Pitch, Formant, __________________________________________________*****_________________________________________________ I. Introduction: The production of speech is a highly complex motor task that involves approximately 100 orofacial, laryngeal , pharyngeal and respiratory muscles . Speech production requires airflow from the lungs (respiration ) to be phonated through the vocal folds of the larynx (phonation ) and resonated in the vocal cavities shaped by the jaw , soft palate , lips , tongue and other articulators (articulation ).It was found that the volume of lungs and breathing pattern during the speech is differed than during the quite respiration and may also be expected to change with different lungs diseases. As a complex motor act, speech requires close neuromuscular coordination of the articulatory process and the phonotory and respiratory process. The latter regulates the subglottal pressure and glottal airflow needed to drive the sound generator. The two functions of the respiratory system can be differentiated as: one provides for the gas exchange necessary for life purpose; other the constant air pressure and air flow required for the production of speech sounds. Speech respiration occurs as long as the primary function, the physiologically required oxygen, and corbondioxide exchange are maintained. Thus the main function of speech respiration is to provide the driving forces necessary for the generation of sounds i.e. to enable the oral communication. Some acoustic voice parameters like fundamental frequencies, Pitch, Intensity, Jitter, Shimmer, maximum phonation time, HNR, NHR and mean autocorrelation can be used for the comparison of normal and affected person. [1]. Chronic obstructive pulmonary disease (COPD) impacts life in many ways. Frequent wheezing and coughing, trouble breathing, coughing up mucus and shortness of breath are just a few COPD symptoms [2]. Also, Vocal cord dysfunction (VCD) is a condition in which the larynx exhibits paradoxical vocal cord adduction during inspiration, resulting in extra- thoracic variable airway obstruction [3]. Bronchial asthma labored breathing & wheezing and allergies can also cause a sore throat & inflammation around the vocal cords. So the voice sound becomes hoarse or scratchy when swollen inflamed cords don‟t vibrate efficiently. A breathing pattern is an oscillatory event having parameters like Force vital capacity (FVC), forced expiratory volume in one second (FEV1) forced expiratory volume in one second/forced vital capacity (FEV1/FVC), peak expiratory flow (PEF). The variation in these parameters from standard value can also lead to the discrimination between healthy and unhealthy person [4]. The present work shows the assessment of respiratory disorders with the help of voice parameters. II. Literature review: Gursimarjot Singh Walia et al [5], developed a method to differentiate five categories of people i) healthy people, ii) people suffering from intermittent asthma, iii) people suffering from mild asthma, iv) people suffering from moderate asthma, and v) people suffering from severe asthma, based on their voice analysis. The method involved the development of a numerical formula using the voice parameters, like Fundamental Frequency, Jitter, Shimmer and Maximum Phonation Time.Thus to determine the Level of Asthma in a person by their voice analysis. Khushboo Batra et al [6] , presented an acoustic analysis of healthy and asthmatic patients voices. Speech recorded for 25 healthy and asthmatic patients between the age of 40 to 65yrs, making them to speak vowels at least five times and the parameters such as jitter, HNR & NHR were taken out and compared for both, in which it was found that Jitter was high for each vowel in case of asthmatic patients and low for healthy persons whereas HNR was high for healthy person and low for asthmatic ones, only vowel „i‟ has no result and NHR was high for asthmatic patients and low for healthy ones. Rachna et al,2014[7],applied a feature extraction technique for extracting similar features in asthmatic patients and normal human being. Mel – Frequency Cepstral Coefficient technique was used for feature extraction process. Thus providing a startup for diagnosing asthma in initial stages, so