Institute for Response-Genetics (e.V.)

Chairman: Prof. Dr. Hans H. Stassen

Psychiatric Hospital (KPPP), University of Zurich

IFRG Logo
IFRG Emblem
Trithemius

COMPARE — Systematic comparisons of spectral voice patterns

Computes the similarity between spectral voice patterns in intra-individual comparisons (repeated assessments on the same individual) and inter-individual comparisons (assessments on different, unrelated/related individuals). The program particularly allows one to analyse the similarity between voice patterns as a function of frequency at a quartertone resolution over the frequency range of 64-8192 Hz, so that frequency bands can be weighted in such a way that the intra-individual similarity is maximized while the inter-individual similarity is minimized. The effects of this optimization can be quantified through the false-negative and false-positive classification errors in a computerized recognition of person test (CRP-test).

            Specificationlist:     COMPARE
            ------------------------------
            I4 PROT                      0  Default-value
            I4 PLOT                      0  Default-value
            I4 PMAX                      0  Default-value
            I4 LPRT                      6  Default-value
            I4 TLOG                      0  Default-value
 
            01 PROT Controls output to display/printer
            02 PLOT Controls graphic output
            03 PMAX Specifies maximum number of plot pages
            04 LPRT Logical unit number of plot-device
            05 TLOG Logarithmic transformation of spectral lines
            06 DEMO Examples that illustrate program function
 
 
            - PROT = 0: No print output
                   = 1: Basic characteristics of spectral voice patterns
                   = 2: Details on optimization
 
            - PLOT = 0: No plot output
                   = 1: Spectral voice patterns
                   = 2: Similarity between spectral voice patterns
 
            - PMAX = q: Maximum number of plot pages
                   = 0: Unlimited number
 
            - LPRT = q: Logical unit number of plot-device (standard=6;
                        valid numbers are 46-96)
 
            - TLOG = 0: No effect
                   = 1: Logarithmic transformation of spectral lines
 
 
            - DEMO: Distribution of Within- and Between-Subject Similarities
 
           

Example

 
            &&START GSELECT=Normative speech study zurich: ages 20-25 (study 600)
             STUD=600,MODE=1,STAT=2,AMIN=20,AMAX=25,PROT=1,RSET=1,PROZ=10
             POPU=AGE20-25
              ?
            &&START GSELECT=Normative speech study zurich: ages 30-35 (study 600)
             AMIN=30,AMAX=35,RSET=0
             POPU=AGE30-35
              ?
            &&START COMPARE=Normative speech study zurich (study 600)
             PROT=1,PLOT=3,LPRT=83
            &&START GSELECT=Normative speech study zurich: ages 20-25 (study 600)
             STUD=600,MODE=2,ATAG=1,AMIN=20,AMAX=25,PROT=1,RSET=1,PROZ=10
             POPU=AGE20-25
            &&START GSELECT=Normative speech study zurich: ages 30-35 (study 600)
             AMIN=30,AMAX=35,RSET=0
             POPU=AGE30-35
            &&START COMPARE=Normative speech study zurich (study 600)
             PROT=1,PLOT=3,LPRT=83
 
           
vSpacer voxFig25
Fig. 25: The voice sound characteristics of male and female speakers as estimated from 97 healthy volunteers (ages 18 - 35 years) with 2 repeated assessments at 14-day intervals display large inter-individual differences but are highly stable over time. This suggests that voice sound characteristics are biologically/genetically predisposed (in the range of 80%). Changes caused by affective disorders are small compared to the highly stable "static" component inherent in human speech.
Hspacer
[ Mail to Webmaster ] k454910@ifrg.ch
rFrame