Evaluation of glottal closure instant detection in a range of voice qualities

Kane, John; Gobl, Christer

doi:10.1016/j.specom.2012.08.011

Cited by 40 publications

(29 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The occurrence of creaky voice was measured using the Voice Analysis Toolkit [24]. The Peak Slope measure [25] calculated from a wavelet-based decomposition of the speech signal into octave bands was used to differentiate breathy and tense voice qualities.…”

Section: Voice Quality Features (Voi)mentioning

confidence: 99%

It Sounds Like You Have a Cold! Testing Voice Features for the Interspeech 2017 Computational Paralinguistics Cold Challenge

Huckvale

Beke

2017

Interspeech 2017

View full text Add to dashboard Cite

This paper describes an evaluation of four different voice feature sets for detecting symptoms of the common cold in speech as part of the Interspeech 2017 Computational Paralinguistics Challenge. The challenge corpus consists of 630 speakers in three partitions, of which approximately one third had a "severe" cold at the time of recording. Success on the task is measured in terms of unweighted average recall of cold/not-cold classification from short extracts of the recordings. In this paper we review previous voice features used for studying changes in health and devise four basic types of features for evaluation: voice quality features, vowel spectra features, modulation spectra features, and spectral distribution features. The evaluation shows that each feature set provides some useful information to the task, with features from the modulation spectrogram being most effective. Feature-level fusion of the feature sets shows small performance improvements on the development test set. We discuss the results in terms of the most suitable features for detecting symptoms of cold and address issues arising from the design of the challenge.

show abstract

Section: Voice Quality Features (Voi)mentioning

confidence: 99%

It Sounds Like You Have a Cold! Testing Voice Features for the Interspeech 2017 Computational Paralinguistics Cold Challenge

Huckvale

Beke

2017

Interspeech 2017

View full text Add to dashboard Cite

show abstract

“…Drugman gives a brief review focused on GCI detection (Thomas Durgman, 2011). In another brief review, the GCI detection over a range of voice qualities was discussed, with emphasis on GCI detection in modal and non-modal phonation (Kane and Gobl, 2013b). The non-modal phonation display varying glottal source characteristics which include voices like creaky, breathy, tense, harsh and falsetto.…”

Section: Glottal Closure Instant Detectionmentioning

confidence: 99%

“…There are several refinements suggested to improve the GCI detection from the LP residual. A recent one is based on filtering the LP residual with a resonator located at approximately at F 0 (Kane and Gobl, 2013b).…”

Section: Glottal Closure Instant Detectionmentioning

confidence: 99%

Glottal source processing: From analysis to applications

Drugman

Alku

Alwan

et al. 2014

Computer Speech & Language

101

View full text Add to dashboard Cite

The great majority of current voice technology applications rely on acoustic features, such as the widely used MFCC or LP parameters, which characterize the vocal tract response. Nonetheless, the major source of excitation, namely the glottal flow, is expected to convey useful complementary information. The glottal flow is the airflow passing through the vocal folds at the glottis. Unfortunately, glottal flow analysis from speech recordings requires specific and complex processing operations, which explains why it has been generally avoided. This paper gives a comprehensive overview of techniques for glottal source processing. Starting from analysis tools for pitch tracking, detection of glottal closure instant, estimation and modeling of glottal flow, this paper discusses how these tools and techniques might be properly integrated in various voice technology applications.

show abstract

“…In a quantitative review of five advanced GCI detection algorithms (Drugman et al 2012), SEDREAMS and ZFR are two best performing techniques in terms of robustness. The SEDREAMS has been later enhanced to better handle voice qualities of different phonation types (Kane and Gobl 2013), but this study only takes the fundamental implementation into account. Both methods start their processing from the speech signal sðnÞ.…”

Section: Performance Evaluationmentioning

confidence: 99%

Robust glottal closure instant detection by jointly exploiting stationary wavelet transform and harmonic superposition

Hsu

2015

Int J Speech Technol

View full text Add to dashboard Cite

This study jointly utilizes stationary wavelet transform (SWT) and harmonic superposition (HS) techniques to locate zero-crossings closely related to glottal closure instants (GCI). The entire process is performed directly on voiced speech signals without referring to the linear prediction residual or voiced source signal derived by inverse filtering. Subsequent to the multi-scale SWT decomposition, a linear phase FIR filter is introduced to translate positive zero-crossings into pulse-like features. While the product across the approximation coefficients in various SWT levels sharpens impulse features, the HS is employed to sieve out the main pulses corresponding GCIs. The advantages of using the proposed SWT-HS scheme for GCI detection are examined using the PTDB-TUG database. Compared with the other two advanced methods, namely the SEDREAMS and ZFR, without the assistance of any refining process the proposed SWT-HS not only renders better accuracy in GCI positioning but exhibits superior robustness against additive noise.

show abstract

Evaluation of glottal closure instant detection in a range of voice qualities

Cited by 40 publications

References 40 publications

It Sounds Like You Have a Cold! Testing Voice Features for the Interspeech 2017 Computational Paralinguistics Cold Challenge

It Sounds Like You Have a Cold! Testing Voice Features for the Interspeech 2017 Computational Paralinguistics Cold Challenge

Glottal source processing: From analysis to applications

Robust glottal closure instant detection by jointly exploiting stationary wavelet transform and harmonic superposition

Contact Info

Product

Resources

About