Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Analyses of voice and glottographic signals in singing and speech
KTH, School of Electrical Engineering and Computer Science (EECS). (TMH)
2018 (English)Doctoral thesis, comprehensive summary (Other academic)
Abstract [en]

Recent advances in machine learning and time series analysis techniques have brought new perspectives to a great number of scientific fields. This thesis contributes applications of such techniques to voice analysis, in an attempt to extract information on the vibration of the vocal folds as such, as well as on the radiated acoustic signal. The data that was analyzed in this work are acoustic recordings, electroglottographic (EGG) signals and transnasal high- speed videoendoscopic images. The data analysis techniques are primarily based on clustering, i.e., grouping of data based on similarity, and sample entropy analysis, i.e., quantifying the degree of irregularity in a given signal. The experiments were conducted so as to provide data for different types of vibratory behaviors (or vibratory states) of the vocal folds. Clustering was used in order to categorize in an unsupervised fashion these different vi- bratory states, based solely on the electroglottographic signal, or the glottal area waveform, or both. Sample entropy was utilized as an indicator of in- stabilities, when subjects produced voiced sounds using irregular vibratory patterns, such as register breaks, intermittent diplophonia, and other types of irregularities. The prominent role of sound pressure level and fundamental frequency motivated further study of the relationship between them and the shape of the electroglottographic waveform. Graphical representations were created to visualize the relationship between different vibratory behaviors with fundamental frequency and sound pressure level. The EGG waveform shape was seen to depend strongly on sound pressure level and somewhat less on fundamental frequency. In very soft phonation, the almost sinusoidal waveform of the EGG suggests that studying the EGG using clusters may give a better representation compared to conventional time-domain metrics. The paradigm of the clustering was later applied in synchronous recordings of electroglottogram and glottal area waveforms in professional tenor singers. Different vibratory states were classified successfully using clustering, and the electroglottogram was seen to be as good as the glottal area waveform for such a classification task. The last part of this work concerns voices from subjects with organic dysphonia. A study was dedicated to investigate how vowel context (sustained versus excerpted from speech) can affect the power of quantitative acoustic measures to discriminate dysphonic subjects from controls. Two acoustic voice quality measures were used: the cepstral peak prominence (smoothed) and sample entropy. The cepstral peak prominence (smoothed) showed better discriminatory power with excerpted vowels, while sample entropy with sustained vowels. Additionally, it was found that sample entropy was strongly correlated with cepstral peak prominence (smoothed) and with the perceptual quality of breathiness. 

Place, publisher, year, edition, pages
Stockholm, Sweden: KTH Royal Institute of Technology, 2018. , p. 55
Series
TRITA-EECS-AVL ; 2018:6
Keywords [en]
voice ; singing ; electroglottography ; clustering ; dysphonia ; sample entropy ;
National Category
Other Natural Sciences
Research subject
Speech and Music Communication
Identifiers
URN: urn:nbn:se:kth:diva-221825ISBN: 978-91-7729-668-3 (print)OAI: oai:DiVA.org:kth-221825DiVA, id: diva2:1177710
Public defence
2018-02-23, F3, Lindstedtsvägen 26, Stockholm, 13:30 (English)
Opponent
Supervisors
Projects
Phonatory dynamics and states
Funder
Swedish Research Council, 2010-4565Swedish Research Council, 2013-0632
Note

QC 20180126

Available from: 2018-01-26 Created: 2018-01-25 Last updated: 2018-01-26Bibliographically approved
List of papers
1. Analysis of vibratory states in phonation using spectral features of the electroglottographic signal
Open this publication in new window or tab >>Analysis of vibratory states in phonation using spectral features of the electroglottographic signal
2014 (English)In: The journal of the Acoustical Society of America, ISSN 0001-4966, Vol. 136, no 5, p. 2773-2783Article in journal (Refereed) Published
Abstract [en]

The vocal folds can oscillate in several different ways, manifest to practitioners and clinicians as ‘registers’ or ‘mechanisms’, of which the two most commonly considered are modal voice and falsetto voice. Here these will be taken as instances of different ‘vibratory states’, i.e., distinct quasi-stationary patterns of vibration of the vocal folds. State transitions are common in biomechanical nonlinear oscillators; and they are often abrupt and impossible to predict exactly. Switching state is much like switching to a different voice. Therefore, vibratory states are a source of confounding variation, for instance, when acquiring a voice range profile (VRP). In the quest for a state-aware, non-invasive VRP, a semi-automatic method based on the short-term spectrum of the electroglottographic signal (EGG) was developed. The method identifies rapid vibratory state transitions, such as the modal-falsetto switch, and clusters the EGG data based on their similarities in the relative levels and phases of the lower frequency components. Productions of known modal and falsetto voice were accurately clustered by a Gaussian mixture model. When mapped into the VRP, this EGG-based clustering revealed connected regions of different vibratory sub-regimes in both modal and falsetto.

Place, publisher, year, edition, pages
Acoustical Society of America (ASA), 2014
Keywords
voice function, phonation, vocal registers, electroglottography, vocal fold vibrations
National Category
Fluid Mechanics and Acoustics
Identifiers
urn:nbn:se:kth:diva-145677 (URN)10.1121/1.4896466 (DOI)000344989000046 ()2-s2.0-84908587626 (Scopus ID)
Projects
FonaDyn
Funder
Swedish Research Council, 2010-4565
Note

Updated from submitted to published.

QC 20140815

Available from: 2014-05-26 Created: 2014-05-26 Last updated: 2018-01-25Bibliographically approved
2. Investigation of the relationship between electroglottogram waveform, fundamental frequency, and sound pressure level using clustering
Open this publication in new window or tab >>Investigation of the relationship between electroglottogram waveform, fundamental frequency, and sound pressure level using clustering
2017 (English)In: Journal of Voice, ISSN 0892-1997, E-ISSN 1873-4588, Vol. 31, no 4, p. 393-400Article in journal (Refereed) Published
Abstract [en]

Although it has been shown in previous research (Orlikoff, 1991; Henrich et al, 2005; Kuang et al, 2014; Awan, 2015) that there exists a relationship between the electroglottogram (EGG) waveform and the acoustic signal, this relationship is still not fully understood. To investigate this relationship, the EGG and acoustic signals were measured for four male amateur choir singers who each produced eight consecutive tones of increasing and decreasing vocal intensity. The EGG signals were processed cycle-synchronously to obtain the discrete Fourier transform, and the data were used as an input to a clustering algorithm. The acoustic signal was analyzed in terms of sound pressure level (dB SPL) and fundamental frequency (f(o)) of vibration, and the results of both EGG and acoustic analysis were depicted on a two-dimensional plane with f(o) on the x-axis and SPL on the y-axis. All the subjects were seen to have a weak, near-sinusoidal EGG waveform in their lowest SPL range, whereas increase in SPL coincided with progressive enrichment in harmonic content of the EGG waveforms. The results of the clustering were additionally used to classify waveforms across subjects to enable inter-subject comparisons and assessment of individual strategies of exploring the f(o)-SPL dimensions. In these male subjects, the EGG waveform shape appeared to vary with SPL and to remain essentially constant with f(o) over one octave.

Place, publisher, year, edition, pages
Institute of Electrical and Electronics Engineers (IEEE), 2017
National Category
Fluid Mechanics and Acoustics
Identifiers
urn:nbn:se:kth:diva-211744 (URN)10.1016/j.jvoice.2016.11.003 (DOI)000406147000001 ()27939138 (PubMedID)2-s2.0-85008154357 (Scopus ID)
Funder
Swedish Research Council, 2010-4565 2013-0642
Note

QC 20170815

Available from: 2017-08-15 Created: 2017-08-15 Last updated: 2018-01-25Bibliographically approved
3. A comparison of electroglottographic and glottal area waveforms for phonation type differentiation in male professional singers
Open this publication in new window or tab >>A comparison of electroglottographic and glottal area waveforms for phonation type differentiation in male professional singers
Show others...
2018 (English)Manuscript (preprint) (Other academic)
Abstract [en]

This study investigates the use of glottographic signals (EGG and GAW) to study phonation in different vibratory states as produced by professionally trained singers. Six western classical tenors were asked to phonate pitch glides from modal to falsetto phonation, or modal to their stage voice above the passaggio (SVaP). For each pitch glide the sample entropy (SampEn) of the EGG signal was calculated to establish a “ground truth” for the performed phonation type; the cycles before the maximum SampEn peak were labeled as modal, and the cycles after the peak as falsetto, or SVaP. Three classifications of vibratory state were performed using clustering: one based only on the EGG, one based on the GAW, and one based on their combi- nation. The classification error rate (clustering vs ground truth) was on average smaller than 10%, for any of the three settings, revealing no special advantage of the GAW over EGG, and vice versa. The EGG-based time domain metric analysis revealed a larger contact quotient and larger normalized EGG derivative peak ratio in modal, compared to SVaP and falsetto. The glottographic waveform comparison of SVaP with falsetto and modal suggests that SVaP resembles more falsetto than modal, though with a larger contact quotient. 

Keywords
classical singing; registers; clustering; electroglottography; glottal area waveform
National Category
Other Natural Sciences
Research subject
Speech and Music Communication
Identifiers
urn:nbn:se:kth:diva-221795 (URN)
Funder
Swedish Research Council, 2010-4565Swedish Research Council, 2013-0642
Note

QC 20180129

Available from: 2018-01-25 Created: 2018-01-25 Last updated: 2018-01-29Bibliographically approved
4. Effect of vowel context in cepstral and entropy analysis of pathological voices
Open this publication in new window or tab >>Effect of vowel context in cepstral and entropy analysis of pathological voices
Show others...
2018 (English)Manuscript (preprint) (Other academic)
Abstract [en]

This study investigates the effect of vowel context (excerpted from speech versus sustained) on two voice quality measures: the cepstral peak prominence smoothed (CPPS) and sample entropy (SampEn). Thirty-one dysphonic subjects with different types of organic dysphonia and thirty- one controls read a phonetically balanced text and phonated sustained [a:] vowels in comfortable pitch and loudness. All the [a:] vowels of the read text were excerpted by automatic speech recognition and phonetic (forced) alignment. CPPS and SampEn were calculated for each subject for the middle 41 ms of each excerpted vowel forming one distribution of CPPS and SampEn values per subject. The middle two seconds of the sustained vowels were analyzed using a 41 ms window, forming another distribution of CPPS and SampEn values per subject. Two speech-language pathologists performed a perceptual evaluation of the dysphonic subjects' voice quality from the recorded text. The power of discriminating the dysphonic group from the controls for SampEn and CPPS was assessed for the excerpted and sustained vowels with the Receiver-Operator Characteristic (ROC) analysis. The best discrimination in terms of Area Under Curve (AUC) for CPPS occurred using the mean of the excerpted vowel distributions (AUC=0.86) and for SampEn using the 95th percentile of the sustained vowel distributions (AUC=0.84). CPPS and SampEn were found to be negatively correlated, and the largest correlation was found between the corresponding 95th percentiles of their distributions (Pearson, r=- 0.83, p < 10-3). A strong correlation was also found between the 95th percentile of SampEn distributions and the perceptual quality of breathiness (Pearson, r=0.83, p < 10-3). The results suggest that depending on the acoustic voice quality measure, sustained vowels can be more effective than excerpted vowels for detecting dysphonia. Additionally, when using CPPS or SampEn there is an advantage of using the measures' distributions rather than their average values. 

Keywords
dysphonia ; voice analysis ; cepstral peak prominence ; sample entropy ; vowel context
National Category
Other Natural Sciences
Research subject
Speech and Music Communication
Identifiers
urn:nbn:se:kth:diva-221797 (URN)
Funder
Swedish Research Council, 2010-4565Swedish Research Council, 2013-0632
Note

QC 20180129

Available from: 2018-01-25 Created: 2018-01-25 Last updated: 2018-01-29Bibliographically approved

Open Access in DiVA

Andreas Selamtzis - 2018- Analyses of voice and glottographic signals in singing and speech(6176 kB)94 downloads
File information
File name FULLTEXT01.pdfFile size 6176 kBChecksum SHA-512
4c97ae9deb9e7e8a4cae04d5e4daed23f328fac1a1089bacea6ca5e5bc8b16868e08ff782cbbe2a2af7f9f976098f6cee73a7a701e42f388c35d09c78d835790
Type fulltextMimetype application/pdf

Search in DiVA

By author/editor
Selamtzis, Andreas
By organisation
School of Electrical Engineering and Computer Science (EECS)
Other Natural Sciences

Search outside of DiVA

GoogleGoogle Scholar
Total: 94 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

isbn
urn-nbn

Altmetric score

isbn
urn-nbn
Total: 20048 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf