Detection of extra pulses in synthesized glottal area waveforms of dysphonic voices

Abstract

Background and objectives The description of production kinematics of dysphonic voices plays an important role in the clinical care of voice disorders. However, high-speed videolaryngoscopy is not routinely used in clinical practice, partly because there is a lack of diagnostic markers that may be obtained from high-speed videos automatically. Aim of the study is to propose and test a procedure that automatically detects extra pulses, which may occur in voiced source signals of pathological voices in addition to cyclic pulses. Material and methods Glottal area waveforms (GAW) are synthesized and used to test a detector for extra pulses. Regarding synthesis, for each GAW a cyclic pulse train is mixed with an extra pulse train, and additive noise. The cyclic pulse trains are varied across GAWs in terms of fundamental frequency, pulse shape, and modulation noise, i.e., jitter and shimmer. The extra pulse trains are varied across GAWs in terms of the height of the extra pulses, and their rates of occurrence. The energy level of the additive noise is also varied. Regarding detection, first, the fundamental frequency is estimated jointly with the cyclic pulse train waveform, second, the modulation noise is estimated, and finally the extra pulse train waveform is estimated. Two versions of the detector are compared, i.e., one that parameterizes the shapes of the cyclic pulses, and one that uses unparameterized pulse shape estimates. Two corpora are used for testing, i.e., one with 100 GAWs containing random extra pulses, and one with 25 GAWs containing extra pulses in the closed phases of each glottal phase representing subharmonic voices. Results and discussion With pulse shape parameterization (PSP) a maximum mean accuracy of 88.3% is achieved when detecting random extra pulses. Without PSP, the maximum mean accuracy reduces to 82.9%. Detection performance decreases if the energy level of additive noise is higher than −25 dB with respect to the energy of the cyclic pulse train, and if the irregularity strength exceeds 0.1. For bicyclic, i.e., subharmonic voices, the approach fails without PSP, whereas with PSP, a mean sensitivity of 87.4% is achieved for subharmonic voices. Conclusion A synthesizer for GAWs containing extra pulses, and a detector for extra pulses are proposed. With PSP, favorable detector performance is observed for not too high levels of additive noise and irregularity strengths. In signals with high noise levels, the detector without PSP outperforms the other one. Detection of extra pulses fails if irregularity strength is large. For subharmonic voices PSP must be used.

How to analyze and manipulate nonlinear phenomena in voice recordings

“…There are numerous proposed solutions for multi-pitch estimation in polyphonic music [91] and conversation analysis [92], but their robustness and applicability to biphonation remain largely unknown. A rare exception is the work by Aichinger, who developed and tested algorithms for simultaneous tracking of two frequencies in pathological human voices, essentially by means of testing several paths through f o candidates [93,94]. Tools for AM analysis may also be applicable when one frequency is many times lower and amplitude-modulates the higher one, but otherwise manual annotation and description based on the spectrogram and the phase space are the only viable approaches at present.…”

Synthesis and Analysis-By-Synthesis of Modulated Diplophonic Glottal Area Waveforms

“…We used an earlier version of MPD for the purpose of detecting extra pulses from 125 synthetic GAWs [27] and for modeling modulation of pulse times and heights in diplophonic pGAWs and audio waveforms [28], where it was tested using 29 clinically observed vowels.…”

“…Ten sequences Q γ are obtained by running Viterbi ten times, and f o candidates f γ o are obtained by mapping Q γ back to the frequency domain, where γ is the candidate index. Instead of selecting f o s and pGAWs via majority voting-based preselection combined with brute force [26], a candidate-wise switching method [27] is adapted here to allow up to two simultaneous f o s. Second, the model is refined by enabling modulations of the pGAW pulse times and prominences by closing the mod-switch shown in Fig. 2b.…”

FATFA-based DRN: Feedback Artificial Tree Firefly Algorithm-Enabled Deep Residual Network for Pathological Speech Enhancement

“…With respect to the perspective on the field of health diagnosis, pathologic voice detection depends on the phonetic indication or physiological defects of the voice scheme. In the medical field, the most commonly used diagnostic methods are Stroboscopy, Laryngoscopy, and Endoscopy, which is considered as a time consuming, persistent, and an invasive method [14] [15]. These techniques require skilled medical experts, well specialized clinical equipment's, and are time consuming and expensive.…”