Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Studies in Musical Acoustics and PsychoacousticsComparison of Vocal and Violin Vibrato with Relationship to the Source/Filter Model

Studies in Musical Acoustics and Psychoacoustics: Comparison of Vocal and Violin Vibrato with... [Although the source/filter model is often mentioned in the literature of acoustics and signal processing (e.g., Gold and Morgan, Speech and Audio Signal Processing, Wiley), it has seldom been implemented for musical instrument sounds. For operatic style male voices with sufficient vibrato depth, overlapped harmonic amplitude-versus-frequency (HAF) graphs can yield displays that trace out vocal tract resonances quite effectively (Maher and Beauchamp in Appl Acoust 30:219–245, 1992 [4]; Arroabarren and Carlosena J Acoust Soc Am 119(4):2483–2497, 2006 [5]). If the glottus signal can be derived, its spectrum (in dB) can be subtracted from the HAF data to reveal a vocal tract transfer function. However, for the violin the HAF method with vibrato excitation has proved unsuccessful because (1) violin vibrato depths are generally insufficient and (2) HAF traces appear too steep to be caused by actual violin resonances. Therefore, a violin glide tone (C5-to-C4, performed in an anechoic chamber) was used instead. Based on an assumption that the source signal spectrum was independent of fundamental frequency (F0), average ratios between adjacent harmonic amplitudes were measured making it possible to derive a source spectrum (within a scale factor). From this the violin transfer function was derived. As a comparison, a pair of violin glide signals (one at the bridge and the other radiated) supplied to the author by Alfonso Perez-Carrillo (J Acoust Soc Am 130(2): 1020–1027, 2011 [19]) were analyzed. The measured bridge spectrum was similar to that of the C5-to-C4 tone’s derived source spectrum, but the derived filter was quite different, as might be expected considering the different violins and arbitrary microphone positions used in the two cases.] http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png

Studies in Musical Acoustics and PsychoacousticsComparison of Vocal and Violin Vibrato with Relationship to the Source/Filter Model

Part of the Current Research in Systematic Musicology Book Series (volume 4)
Editors: Schneider, Albrecht
Beauchamp, James W.
Studies in Musical Acoustics and Psychoacoustics — Dec 27, 2016
Download PDF
20 pages

Loading next page...
 
/lp/springer-journals/studies-in-musical-acoustics-and-psychoacoustics-comparison-of-vocal-KpOGkmXfjt
Publisher
Springer International Publishing
Copyright
© Springer International Publishing AG 2017
ISBN
978-3-319-47291-1
Pages
201 –221
DOI
10.1007/978-3-319-47292-8_7
Publisher site
See Chapter on Publisher Site

Abstract

[Although the source/filter model is often mentioned in the literature of acoustics and signal processing (e.g., Gold and Morgan, Speech and Audio Signal Processing, Wiley), it has seldom been implemented for musical instrument sounds. For operatic style male voices with sufficient vibrato depth, overlapped harmonic amplitude-versus-frequency (HAF) graphs can yield displays that trace out vocal tract resonances quite effectively (Maher and Beauchamp in Appl Acoust 30:219–245, 1992 [4]; Arroabarren and Carlosena J Acoust Soc Am 119(4):2483–2497, 2006 [5]). If the glottus signal can be derived, its spectrum (in dB) can be subtracted from the HAF data to reveal a vocal tract transfer function. However, for the violin the HAF method with vibrato excitation has proved unsuccessful because (1) violin vibrato depths are generally insufficient and (2) HAF traces appear too steep to be caused by actual violin resonances. Therefore, a violin glide tone (C5-to-C4, performed in an anechoic chamber) was used instead. Based on an assumption that the source signal spectrum was independent of fundamental frequency (F0), average ratios between adjacent harmonic amplitudes were measured making it possible to derive a source spectrum (within a scale factor). From this the violin transfer function was derived. As a comparison, a pair of violin glide signals (one at the bridge and the other radiated) supplied to the author by Alfonso Perez-Carrillo (J Acoust Soc Am 130(2): 1020–1027, 2011 [19]) were analyzed. The measured bridge spectrum was similar to that of the C5-to-C4 tone’s derived source spectrum, but the derived filter was quite different, as might be expected considering the different violins and arbitrary microphone positions used in the two cases.]

Published: Dec 27, 2016

Keywords: Vocal Tract; Source Spectrum; Harmonic Amplitude; Spectral Envelope; Anechoic Chamber

There are no references for this article.