USDOE Human Subjects Research Database, Fiscal Year 2001

Argonne National Laboratory

Public Information Contact:

Mr. David M. Baurac
Argonne National Laboratory
9700 S. Cass Avenue
Argonne, IL 60439-4833

Phone: 630-252-5584
Fax: 630-252-5274
E-mail: baurac@anl.gov

Institutional Review Board (IRB):

Projects are approved by an IRB located at: Argonne National Laboratory
The approving IRB operates under a Multiple Project Assurance (MPA), Single Project Assurance (SPA), or Cooperative Project Assurance (CPA) recognized by DOE or by the Department of Health and Human Services (HHS).
MPA number of the IRB: DOE.SPA.ANL

Human Subject Projects:

Number of Human Subjects projects reported: 1

ANL-00-00001: "Speaker Recognition Through NLP and CWT Modeling"

Go to Human Subjects Research 2001 main page

Project Identifier: ANL-00-00001

Project Title:

"Speaker Recognition Through NLP and CWT Modeling"

Principal Investigator: Dr. Stefania A. Brown-VanHoozer, Argonne National Laboratory

Project started in: 2000
This project ended in fiscal year 2001.

Project Funding Information:

This project did not receive funding during fiscal year 2001.
This project did not use human subjects in fiscal year 2001.
Explanation: Activities involving human subjects were completed in FY2000. Subsequent work was restricted to the final analysis of data.

Funding for Human Subjects Research: No Funding Sources Reported

Information on Use of Human Subjects:

This project does not involve the use of multiple protocols/subprojects.

Institutional Review Board (IRB) Review:
Type of Review: Full Board
Approving Institution: Argonne National Laboratory
Most recent approval: 11/10/00
IRB approval number: ANL-00/001

Number of human subjects who participated in this project/protocol/subproject in the last reporting period: 0
Reporting period for number of human subjects: Fiscal Year 2001

Type(s) of Human Subjects Involvement:

Use of personally identifiable data from questionnaires, surveys, or epidemiological studies: Using data collected from subjects specifically for this project.

Use of human beings to develop/test instruments, materials, devices, objects, and the like or to investigate the "man-machine interface".

Abstract:

(a. Objectives, b. Methodology, c. Ionizing Radiation, Radioactive Substances, or Chemical Substances to which human subjects are exposed, d. Involvement of Human Subjects [d.1. procedures used, d.2. risks if any])

Objectives and Methodology

The objective of the research, (speaker recognition through neural-linguistic programming and continuous wavelet modeling), was to develop a method for automatically analyzing voice patterns which could potentially be used in the development of advanced voice recognition systems.

The current technology does not take into account key voice characteristics (e.g., tempo, tone, pitch, etc.) of a person's speech pattern. Thus, the hypothesis proposed in this research is the following: Since individuals have an internal preference for processing information--either visually (by sight), auditorily (by sound) or kinestheitcally (through emotions, tactile sensations, smells or tastes)--we can utilize this to categorize individuals and compare physiological characteristics of the speech patterns within each category. For example, it is proposed that individuals who prefer to internalize information visually (by sight) will have similar physiological characteristics in their speech patterns, and consequently, the same for those individuals whose preference is auditor or kinesthetic.

It is the intent of this study to establish a baseline from a control group consisting of different individuals (20-40 subjects) whose primary representation system (visual, auditory or kinesthetic) can be established using non-verbal cues and verbal predicates. Once all subjects have been categorized, then distinguishing similarities in their speech patterns will be recorded. Results of the data will be mathematically modeled for software coding and tested for feasibility using the TIMIT AND NTIMIT databases. (The Massachusetts Institute of Technology (MIT) and the National Institute of Standards and Technology (NIST) generated the databases that consist of 630 speakers uttering 10 phrases each. The databases are accessible for public use.)

Involvement of Human Subjects
Each individual who volunteered to participate in the research project is interviewed and videotaped to determine how they process information (visually, auditorily, or kinesthetically) based on the parameters of the methodology, Neuro-Linguistic Programming. Following the interview, each subject's voice is recorded while (a) reading a prewritten script, (b) saying (or repeating) a sentence, phrase(s), or a string of characters without a prewritten script, (c) saying a few sentences on a topic of their choice. The subjects are not asked to say or read anything vulgar or obscene, or that in anyway may make them uncomfortable.

The risk factors to the subjects are minimal. The risk for physical injury is similar to that presented by day-to-day activities in an office environment.

An informed consent form is read and signed by each participant, and a copy of the form is provided to the individual. In addition, the research study is verbally described to each subject before and if necessary, during the study. The subjects are allowed to ask questions at any time during the research study, and to refuse to participate or end participation at any time. They also have the right to ask that any data involving them be erased, even after the testing was complete. All material regarding human subjects is stored in a locked area when not in use. The material cannot be analyzed by anyone other than the stated investigator in the consent form without the written permission of the subject.

Go to list of projects at Argonne National Laboratory