A study to collect voice samples of patients and healthy individuals having similar age.

Completed

Conditions: Depression, Asthma, Parkinsonâ€™s disease, Alzheimerâ€™s disease, CongestiveHeart Failure, Pregnancy and Maternal Health, and any other disease condition; and age matched healthy people

Brief Summary: The content and quality of vocal production may be a valuable marker of health information. Vocal

production, for example speech, contains rich information in both the content (linguistic) and quality (non-linguistic)

characteristics of sound information.

Previous research has used vocal samples to predict important markers of cognitive and affective state. From the

non-linguistic information of speech, including but not limited to the information about the pitch, phonemes, and

power spectrum of the vocal signal, researchers have been able to predict with high accuracy the cognitive load of

an individual at a given moment.We propose to use speech and sound data collected from mobile devices to obtain indices of mental and physical

health. We will collect prompted and free speech from a mobile application, including several different cognitive

tasks (eg. reading a sentence or passage, naming colors) and will collect demographic and questionnaire data. We

will then use a proprietary back-end analysis to extract features and correlate these features to cognitive and

affective measures.

Recruitment & Eligibility

Inclusion Criteria

1.All individuals over 18 years of age 2.
Equal numbers of male (3,000) (subject to extension)and female (3,000)(subject to extension) participants 3.
Willing and consenting participants 4.
Potential participant should be able to read.

Exclusion Criteria

Primary Outcome Measures

Name	Time	Method
To examine linguistic and non-linguistic vocal characteristics in diseased and general population	As it is a survey, voice samples will be collected from the patients on mobile base app, only once, i.e. at the time of enrollment. \| After that collected data will be analysed after completion of the study.
ï‚•ï€ To assess the feasibility of collecting large-scale, data-rich samples of voice and sound and of extracting linguistic and non-linguistic information from speech and sound samples recorded over a mobile device,	As it is a survey, voice samples will be collected from the patients on mobile base app, only once, i.e. at the time of enrollment. \| After that collected data will be analysed after completion of the study.
ï‚•ï€ To determine the feasibility of using extracted information to predict demographic (eg. age, gender) and health measures.	As it is a survey, voice samples will be collected from the patients on mobile base app, only once, i.e. at the time of enrollment. \| After that collected data will be analysed after completion of the study.

Secondary Outcome Measures

Name	Time	Method
ï‚•ï€ To assess the feasibility of collecting large-scale, data-rich samples of voice and sound and of extracting linguistic and non-linguistic information from speech and sound samples recorded over a mobile device,	ï‚•ï€ To determine the feasibility of using extracted information to predict demographic (eg. age, gender) and health measures.

Locations (5): Ayurved Rugnalaya &Sterling Multispeciality Hospital
🇮🇳
Pune, MAHARASHTRA, India
Chintamani Hospital
🇮🇳
Pune, MAHARASHTRA, India
Dhanvantari Hospital
🇮🇳
Pune, MAHARASHTRA, India
Dr. Kulkarni Clinic
🇮🇳
Pune, MAHARASHTRA, India
Surya Prabha Nursing Home
🇮🇳
Pune, MAHARASHTRA, India
Ayurved Rugnalaya &Sterling Multispeciality Hospital
🇮🇳Pune, MAHARASHTRA, India
Dr Medha Joshi
Principal investigator
9881465798
medhamaheshjoshi@gmail.com