Quantifying the speaking voice: Generating a speaker code as a means of speaker identification using a simple code-matching technique

Peter Popolo; Richard W. Sanders; Ingo R. Titze

Quantifying the speaking voice: Generating a speaker code as a means of speaker identification using a simple code-matching technique

Peter Popolo, Richard W. Sanders, Ingo R. Titze

Communication Sciences and Disorders

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

2 Scopus citations

Abstract

This paper looks at a methodology of quantifying the speaking voice, by which temporal and spectral features of the voice are extracted and processed to create a numeric code that identifies speakers, so those speakers can be searched in a database much like fingerprints. The parameters studied include: (1) average fundamental frequency (F0) of the speech signal over time, (2) standard deviation of the F0, (3) the slope and (4) sign of the FO contour, (5) the average energy, (6) the standard deviation of the energy, (7) the spectral energy contained from 50 Hz to 1,000 Hz, (8) the spectral energy from 1,000 Hz to 5,000 Hz, (9) the Alpha Ratio, (10) the average speaking rate, and (11) the total duration of the spoken sentence.

Original language	English
Title of host publication	Audio Engineering Society - 123rd Audio Engineering Society Convention 2007
Pages	456-468
Number of pages	13
Volume	1
State	Published - 1 Dec 2007
Event	123rd Audio Engineering Society Convention 2007 - New York, NY, United States Duration: 5 Oct 2007 → 8 Oct 2007

Other

Other	123rd Audio Engineering Society Convention 2007
Country/Territory	United States
City	New York, NY
Period	5/10/07 → 8/10/07

Cite this

@inproceedings{87bed64431de4077b0a0b8b5e546fc02,

title = "Quantifying the speaking voice: Generating a speaker code as a means of speaker identification using a simple code-matching technique",

abstract = "This paper looks at a methodology of quantifying the speaking voice, by which temporal and spectral features of the voice are extracted and processed to create a numeric code that identifies speakers, so those speakers can be searched in a database much like fingerprints. The parameters studied include: (1) average fundamental frequency (F0) of the speech signal over time, (2) standard deviation of the F0, (3) the slope and (4) sign of the FO contour, (5) the average energy, (6) the standard deviation of the energy, (7) the spectral energy contained from 50 Hz to 1,000 Hz, (8) the spectral energy from 1,000 Hz to 5,000 Hz, (9) the Alpha Ratio, (10) the average speaking rate, and (11) the total duration of the spoken sentence.",

author = "Peter Popolo and Sanders, {Richard W.} and Titze, {Ingo R.}",

year = "2007",

month = dec,

day = "1",

language = "English",

isbn = "9781604239027",

volume = "1",

pages = "456--468",

booktitle = "Audio Engineering Society - 123rd Audio Engineering Society Convention 2007",

note = "123rd Audio Engineering Society Convention 2007 ; Conference date: 05-10-2007 Through 08-10-2007",

}

Quantifying the speaking voice: Generating a speaker code as a means of speaker identification using a simple code-matching technique. / Popolo, Peter; Sanders, Richard W.; Titze, Ingo R.
Audio Engineering Society - 123rd Audio Engineering Society Convention 2007. Vol. 1 2007. p. 456-468.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Quantifying the speaking voice

T2 - 123rd Audio Engineering Society Convention 2007

AU - Popolo, Peter

AU - Sanders, Richard W.

AU - Titze, Ingo R.

PY - 2007/12/1

Y1 - 2007/12/1

N2 - This paper looks at a methodology of quantifying the speaking voice, by which temporal and spectral features of the voice are extracted and processed to create a numeric code that identifies speakers, so those speakers can be searched in a database much like fingerprints. The parameters studied include: (1) average fundamental frequency (F0) of the speech signal over time, (2) standard deviation of the F0, (3) the slope and (4) sign of the FO contour, (5) the average energy, (6) the standard deviation of the energy, (7) the spectral energy contained from 50 Hz to 1,000 Hz, (8) the spectral energy from 1,000 Hz to 5,000 Hz, (9) the Alpha Ratio, (10) the average speaking rate, and (11) the total duration of the spoken sentence.

AB - This paper looks at a methodology of quantifying the speaking voice, by which temporal and spectral features of the voice are extracted and processed to create a numeric code that identifies speakers, so those speakers can be searched in a database much like fingerprints. The parameters studied include: (1) average fundamental frequency (F0) of the speech signal over time, (2) standard deviation of the F0, (3) the slope and (4) sign of the FO contour, (5) the average energy, (6) the standard deviation of the energy, (7) the spectral energy contained from 50 Hz to 1,000 Hz, (8) the spectral energy from 1,000 Hz to 5,000 Hz, (9) the Alpha Ratio, (10) the average speaking rate, and (11) the total duration of the spoken sentence.

UR - http://www.scopus.com/inward/record.url?scp=84866495115&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84866495115

SN - 9781604239027

VL - 1

SP - 456

EP - 468

BT - Audio Engineering Society - 123rd Audio Engineering Society Convention 2007

Y2 - 5 October 2007 through 8 October 2007

ER -

Quantifying the speaking voice: Generating a speaker code as a means of speaker identification using a simple code-matching technique

Abstract

Other

Other files and links

Fingerprint

Cite this