Search

Sound generation during voiced speech remains an open research topic because the underlying process within the human larynx is hardly accessible for direct measurements. In the present study, harmonic sound generation during phonation was investigated with a model that replicates the fully coupled fluid-structure-acoustic interaction (FSAI). The FSAI was captured using a multi-modal approach by measuring the flow and acoustic source fields based on particle image velocimetry, as well as the surface velocity of the vocal folds based on laser vibrometry and high-speed imaging. Strong harmonic sources were localized near the glottis, as well as further downstream, during the presence of the supraglottal jet. The strongest harmonic content of the vocal fold surface motion was verified for the area near the glottis, which directly interacts with the glottal jet flow. Also, the acoustic back-coupling of the formant frequencies onto the harmonic oscillation of the vocal folds was verified. These findings verify that harmonic sound generation is the result of a strong interrelation between the vocal fold motion, modulated flow field, and vocal tract geometry.

Background In the CoVID-19 pandemic, singing came into focus as a high-risk activity for the infection with airborne viruses and was therefore forbidden by many governmental administrations. Objective The aim of this study is to investigate the effectiveness of surgical masks regarding the spatial and temporal dispersion of aerosol and droplets during professional singing. Methods Ten professional singers performed a passage of the Ludwig van Beethoven’s “Ode of Joy” in two experimental setups—each with and without surgical masks. First, they sang with previously inhaled vapor of e-cigarettes. The emitted cloud was recorded by three cameras to measure its dispersion dynamics. Secondly, the naturally expelled larger droplets were illuminated by a laser light sheet and recorded by a high-speed camera. Results The exhaled vapor aerosols were decelerated and deflected by the mask and stayed in the singer’s near-field around and above their heads. In contrast, without mask, the aerosols spread widely reaching distances up to 1.3 m. The larger droplets were reduced by up to 86% with a surgical mask worn. Significance The study shows that surgical masks display an effective tool to reduce the range of aerosol dispersion during singing. In combination with an appropriate aeration strategy for aerosol removal, choir singers could be positioned in a more compact assembly without contaminating neighboring singers all singers.

The presented filtering technique is proposed to detect errors and correct outliers inside the acoustic sources, respectively, the first time derivative of the incompressible pressure obtained from large eddy simulations with prescribed vocal fold motion using overlay mesh methods. Regarding the perturbed convective wave equation, the time derivative of the incompressible pressure is the primary sound source in the human phonation process. However, the incompressible pressure can be erroneous and have outliers when fulfilling the divergence-free constraint of the velocity field. This error is primarily occurring for non-conserving prescribed vocal fold motions. Therefore, the method based on a continuous stationary random process was designed to detect rare events in the time derivative of the pressure. The detected events are then localized and treated by a defined window function to increase their probability. As a consequence, the data quality of the non-linearly filtered data is enhanced significantly. Furthermore, the proposed method can also be used to assess convergence of the aeroacoustic source terms, and detect regions and time intervals, which show a non-converging behavior by an impulse-like structure.

Purpose High-speed videoendoscopy (HSV) is an emerging, but barely used, endoscopy technique in the clinic to assess and diagnose voice disorders because of the lack of dedicated software to analyze the data. HSV allows to quantify the vocal fold oscillations by segmenting the glottal area. This challenging task has been tackled by various studies; however, the proposed approaches are mostly limited and not suitable for daily clinical routine. Method We developed a user-friendly software in C# that allows the editing, motion correction, segmentation, and quantitative analysis of HSV data. We further provide pretrained deep neural networks for fully automatic glottis segmentation. Results We freely provide our software Glottis Analysis Tools (GAT). Using GAT, we provide a general threshold-based region growing platform that enables the user to analyze data from various sources, such as in vivo recordings, ex vivo recordings, and high-speed footage of artificial vocal folds. Additionally, especially for in vivo recordings, we provide three robust neural networks at various speed and quality settings to allow a fully automatic glottis segmentation needed for application by untrained personnel. GAT further evaluates video and audio data in parallel and is able to extract various features from the video data, among others the glottal area waveform, that is, the changing glottal area over time. In total, GAT provides 79 unique quantitative analysis parameters for video- and audio-based signals. Many of these parameters have already been shown to reflect voice disorders, highlighting the clinical importance and usefulness of the GAT software. Conclusion GAT is a unique tool to process HSV and audio data to determine quantitative, clinically relevant parameters for research, diagnosis, and treatment of laryngeal disorders. Supplemental Material https://doi.org/10.23641/asha.14575533

In recent studies, it has been assumed that vocal tract formants (Fn) and the voice source could interact. However, there are only few studies analyzing this assumption in vivo. Here, the vowel transition /i/-/a/-/u/-/i/ of 12 professional classical singers (6 females, 6 males) when phonating on the pitch D4 [fundamental frequency (ƒo) ca. 294 Hz] were analyzed using transnasal high speed videoendoscopy (20.000 fps), electroglottography (EGG), and audio recordings. Fn data were calculated using a cepstral method. Source-filter interaction candidates (SFICs) were determined by (a) algorithmic detection of major intersections of Fn/nƒo and (b) perceptual assessment of the EGG signal. Although the open quotient showed some increase for the /i-a/ and /u-i/ transitions, there were no clear effects at the expected Fn/nƒo intersections. In contrast, ƒo adjustments and changes in the phonovibrogram occurred at perceptually derived SFICs, suggesting level-two interactions. In some cases, these were constituted by intersections between higher nƒo and Fn. The presented data partially corroborates that vowel transitions may result in level-two interactions also in professional singers. However, the lack of systematically detectable effects suggests either the absence of a strong interaction or existence of confounding factors, which may potentially counterbalance the level-two-interactions.

Group singing has been associated with higher transmission risks via exhaled and spread aerosols in the CoVID19 pandemic. For this reason, many musical activities, such as rehearsals and lessons, but also voice therapy sessions, have been restricted in many countries. Consequently, transmission risks and pathways have been studied, such as aerosol amounts generated by exhalation tasks, convectional flows in rooms, or the impulse dispersion of different kinds of phonation. The use of water resistance exercises such as those utilizing LAX VOX®, are common in voice lessons and as vocal warm-ups. With this context, this study investigates the impulse dispersion characteristics of aerosols during a voiced water resistance exercise in comparison to normal singing.Twelve professional singers (six male, six female) were asked to phonate a stable pitch through a silicone tube into a bottle filled with water, holding the end of the tube 5 cm below the surface. Before performing the tasks, the singers inhaled the vapor consisting of 0.5 L base liquid from an e-cigarette. The exhaled gas cloud coming out of the bottle was recorded in all three spatial directions and the dispersion was measured as a function of time.At the end of the phonation task, the median distance to the front was 0.55 m and the median of the lateral expansion of the cloud was 0.89 m, the maximum to the front reached 0.88 m, and the maximum of lateral expansion 1.05 m. For the upwards direction of the clouds a median of 1.00 m and a maximum of 1.34 m from the mouth were measured. Three seconds after the end of the task, the medians were declining.The exhaled aerosol cloud can expand despite the obstacle of the water when using LAX VOX® during phonation.

The process of voice production is a complex process and depends on the correct interaction of the vocal folds and the glottal airstream inducing the primary voice source, which is subsequently modulated by the vocal tract. Due to the restricted access to the glottis, not all aspects of the three-dimensional process can be captured by measurements without influencing the measurement object. Hence, the application of a numerical tool capturing the physical process of phonation can provide an extended database for voice treatment and, therefore, can contribute to an increased effectiveness of voice treatment. However, such numerical models involve complex and demanding procedures to model the material behavior and the mechanical contact of the vocal folds and to realize moving boundaries of the involved physical domains. The present paper proposes a numerical model called simVoice, which circumvents these computational expenses by prescribing the experimentally obtained vocal fold motion within the simulation. Additionally, a hybrid approach for sound computation further enhances the computational efficiency and yields good agreement with acoustic measurements. An analysis of the computational workloads suggests that the key factor for a further increase in efficiency is an optimized flow simulation and source term computation.

With age, the atrophy of the thyroarytenoid muscle (TAM), and thus atrophy of the vocal folds, leads to decreased glottal closure, increased breathiness, and a loss in voice quality, which results in a reduced quality of life. A method to counteract the atrophy of the TAM is to induce hypertrophy in the muscle by functional electric stimulation (FES). In this study, phonation experiments were performed with ex vivo larynges of six stimulated and six unstimulated ten-year-old sheep to investigate the impact of FES on phonation. Electrodes were implanted bilaterally near the cricothyroid joint. FES treatment was provided for nine weeks before harvesting. The multimodal measurement setup simultaneously recorded high-speed video of the vocal fold oscillation, the supraglottal acoustic signal, and the subglottal pressure signal. Results of 683 measurements show a 65.6% lower glottal gap index, a 22.7% higher tissue flexibility (measured by the amplitude to length ratio), and a 473.7% higher coefficient of determination (R2) of the regression of subglottal and supraglottal cepstral peak prominence during phonation for the stimulated group. These results suggest that FES improves the phonatory process for aged larynges or presbyphonia.

Glottal closure insufficiency (GCI) and asymmetric oscillations of the vocal folds are often responsible for low voice quality and increased phonation effort for concerned patients. In this study, four clinically observed types of GCI with increasing glottal closure insufficiency were reproduced in the aero-acoustic computer model simVoice. Each type was simulated with symmetric and asymmetric oscillations extracted from high-speed recordings obtained in ex-vivo porcine larynges. The results show a permanent glottal flow throughout the oscillation cycle and an increasing mean flow rate for increasing minimum glottal gap. Consequently, the aerodynamic work exerted onto the vocal folds decreases for an increasing gap corresponding to the reported patient’s effort increase. Moreover, the quality of the computed sound decreases with an increasing gap based on the cepstral-peak-prominence. Even worse, asymmetric oscillations further degrade both, the aerodynamic energy transfer and the acoustic quality. The temporal evolution of energy transfer shows that the main decay occurs during the opening of the glottis. This shows that the blockage of the flow during glottis closure and the resulting large temporal gradients in pressure and velocity in the larynx are the key features for an efficient phonation process which is also visible in the aero-acoustic sound sources.

Understanding the risk of infection by routine medical examination is important for the protection of the medical personnel. In this study we investigated respiratory particles emitted by patients during routine otolaryngologic procedures and assessed the risks for the performing physician. We developed two experimental setups to measure aerosol and droplet emission during rigid/flexible laryngoscopy, rhinoscopy, pharyngoscopy, otoscopy, sonography and patient interview for subjects with and without masks. A high-speed-camera setup was used to detect ballistic droplets (approx. > 100 µm) and an aerosol-particle-sizer was used to detect aerosol particles in the range of 0.3 µm to 10 µm. Aerosol particle counts were highly increased for coughing and slightly increased for heavy breathing in subjects without masks. The highest aerosol particle counts occurred during rigid laryngoscopy. During laryngoscopy and rhinoscopy, the examiner was exposed to increased particle emission due to close proximity to the patient’s face and provoked events such as coughing. However, even during sonography or otoscopy without a mask, aerosol particles were expelled close to the examiner. The physician’s exposure to respiratory particles can be reduced by deliberate choice of examination technique depending on medical indication and the use of appropriate equipment for the examiners and the patients (e.g., FFP2 masks for both).

IntroductionGroup musical activities using wind instruments have been restricted during the CoVID19 pandemic due to suspected higher risk of virus transmission. It was presumed that the aerosols exhaled through the tubes while playing would be ejected over larger distances and spread into the room due to jet stream effects. In particular, the soprano recorder is widely used as instrument in school classes and for beginners of all age groups in their musical education, as well as in contexts of leisure activities and in professional concert performances.Understanding the aerosol impulse dispersion characteristics of playing the soprano recorder could assist with the establishment of concepts for safe music-making.MethodsFive adult professionally trained soprano recorder players (4 female, 1 male) played four bars of the main theme of L. van Beethoven’s “Ode to Joy” in low and in high octaves, as well as with 3 different potential protection devices in the high octave. For comparison they spoke the corresponding text by F. Schiller. Before each task, they inhaled .5 L of vapor from an e-cigarette filled with base liquid. The vapor cloud escaping during speaking or playing was recorded by cameras and its spread was measured as a function of time in the three spatial dimensions. The potential safety devices were rated in practicability with a questionnaire, and their influence on the sound was compared, generating a long-term average spectrum from the audio data.ResultsWhen playing in the high octave, at the end of the task the clouds showed a median distance of 1.06 m to the front and .57 m diameter laterally (maxima: x: 1.35 m and y: .97 m). It was found that the clouds’ expansion values in playing the recorder with and without safety measures are mostly lower when compared to the ordinary, raised speaking voice of the same subjects.The safety devices which covered the instrument did not show clear advantages and were rated as unpractical by the subjects. The most effective reduction of the cloud was reached when playing into a suction funnel.ConclusionThe aerosol dispersion characteristics of soprano recorders seem comparable to clarinets. The tested safety devices which covered holes of the instrument did not show clear benefits.

In current voice research, there is a growing interest in semi-occluded vocal tract exercises and their contribution to an increased vocal efficiency. However, there are only few studies on the lasting effects.Eight vocally healthy subjects were asked to sustain a phonation at a comfortable loudness on the vowel /i/, with a Fundamental Frequency of 250 Hz (females) or 125 Hz (males). During phonation the subjects were simultaneously recorded with transnasal High Speed Videoendoscopy (HSV, 20,000fps), electroglottography, and audio signals. The subjects then performed a water resistance therapy (WRT) for 10 min (tube: 30 cm length, 5 cm below water surface, diameter 9 mm). Repeated measurements of sustained phonation were performed 0, 5, 10, 20, and 30 min after exercising. From the HSV material the Glottal Area Waveform (GAW) was segmented and GAW parameters were computed.There were strong inter-individual differences concerning the changes of different measures over time after WRT. In general, directly after WRT there was a lowering of the GAW derived Period Perturbation Quotient, a lowering of the Closing Quotient and an increase of the Sound Pressure Level (SPL) in comparison to the pre-intervention measurement. However, only 5 min post WRT there was no longer a clear difference compared to baseline results before WRT. Other values such as Open Quotients exhibited no evident changes by WRT at any timepoint.WRT showed strong inter-individual effects concerning the changes in the evaluated measures post-intervention. General tendencies of some measures directly after the intervention showed a brief effect of only a few minutes.

Introduction Vocal loading capacity is an important aspect of vocal health and is measured using standardized vocal loading tests. However, it remains unclear how vocal fold oscillation patterns are influenced by a standardized vocal loading task. Methods 21 (10 male, 11 female) vocally healthy subjects were analyzed concerning the dysphonia severity index (DSI) and high speed videolaryngoscopy (HSV) on the vowel /i/ at a comfortable pitch and loudness before and after a standardized vocal loading test (10 min standardized text reading, at a level higher than 80 dB (A) measured at 30 cm from the mouth). Results Changes in DSI were statistically significant, diminishing by 1.2 points after the vocal loading test, which was mainly caused by an increase of the minimum intensity. However, the pre-post comparison of HSV derived measures failed to show any statistically significant changes. Conclusion It seems necessary to analyze the effects of a standardized vocal loading test on vocal fold oscillation patterns with respect to softest phonation and phonation threshold pressure rather than comfortable pitch and loudness. Level of evidence 2c

A hybrid aeroacoustic approach was developed for the efficient numerical computation of human phonation. In the first step, an incompressible flow simulation on a three-dimensional (3 D) computational grid, which is capable of resolving all relevant turbulent scales, is performed using STARCCM+ and finite volume method. In the second step, the acoustic source terms on the flow grid are computed and a conservative interpolation to the acoustic grid is performed. Finally, the perturbed convective wave equation is solved to obtain the acoustic field in 3 D with the finite element solver CFS++. Thereby, the conservative transformation of the acoustic sources from the flow grid to the acoustic grid is a key step to allow coarse acoustic grids without reducing accuracy. For this transformation, two different interpolation strategies are compared and grid convergence is assessed. Overall, 16 simulation setups are compared. The initial (267 000 degrees of freedom) and the optimized (21 265 degrees of freedom) simulation setup were validated by measurements of a synthetic larynx model. To conclude, the total computational time of the acoustic simulation is reduced by 95% compared to the initial simulation setup without a significant reduction of accuracy, being 7%, in the frequency range of interest.

Background: Various voice assessment tools, such as questionnaires and aerodynamic voice characteristics, can be used to assess vocal function of individuals. However, not much is known about the best combinations of these parameters in identification of functional dysphonia in clinical settings. Methods: This study investigated six scores from clinically commonly used questionnaires and seven acoustic parameters. 514 females and 277 males were analyzed. The subjects were divided into three groups: one healthy group (N01) (49 females, 50 males) and two disordered groups with perceptually hoarse (FD23) (220 females, 96 males) and perceptually not hoarse (FD01) (245 females, 131 males) sounding voices. A tree stumps Adaboost approach was applied to find the subset of parameters that best separates the groups. Subsequently, it was determined if this parameter subset reflects treatment outcome for 120 female and 51 male patients by pairwise pre- and post-treatment comparisons of parameters. Results: The questionnaire “Voice-related-quality-of-Life” and three objective parameters (“maximum fundamental frequency”, “maximum Intensity” and “Jitter Percent”) were sufficient to separate the groups (accuracy ranging from 0.690 (FD01 vs. FD23, females) to 0.961 (N01 vs. FD23, females)). Our study suggests that a reduced parameter subset (4 out of 13) is sufficient to separate these three groups. All parameters reflected treatment outcome for patients with hoarse voices, Voice-related-quality-of-Life showed improvement for the not hoarse group (FD01). Conclusion: Results show that single parameters are insufficient to separate voice disorders but a set of several well-chosen parameters is. These findings will help to optimize and reduce clinical assessment time., In this study, we show that only a small set of four parameters is sufficient to differentiate between healthy subjects and patients with diagnosis of functional dysphonia. These four parameters also reflect treatment outcome for functional dysphonia and perceivable hoarse voices (success or failure). Thus, in future a functioning set of objective tools could be created that improves and accelerates assessment and therapy of voice disorders.

Acoustic assessment of voice impairment is commonly performed by subjective perceptual evaluation of continuous speech based on a grading system such as the roughness, breathiness, hoarseness (RBH) scale. Here, we present an automatic approach to objectively quantify hoarseness severity based on sustained vowels. For this study, a total of 635 recordings of the sustained vowel /a/ were collected. Temporal, spectral and cepstral features were extracted from one second of each recording. All recordings have an assigned RBH value, which was determined subjectively by an expert based on continuous speech of the respective subject. In order to account for the label noise introduced by different valuation bases, subjects were divided into two levels of hoarseness H

Endoscopic high-speed video (HSV) systems for visualization and assessment of vocal fold dynamics in the larynx are diverse and technically advancing. To consider resulting “concepts shifts” for neural network (NN)-based image processing, re-training of already trained and used NNs is necessary to allow for sufficiently accurate image processing for new recording modalities. We propose and discuss several re-training approaches for convolutional neural networks (CNN) being used for HSV image segmentation. Our baseline CNN was trained on the BAGLS data set (58,750 images). The new BAGLS-RT data set consists of additional 21,050 images from previously unused HSV systems, light sources, and different spatial resolutions. Results showed that increasing data diversity by means of preprocessing already improves the segmentation accuracy (mIoU + 6.35%). Subsequent re-training further increases segmentation performance (mIoU + 2.81%). For re-training, finetuning with dynamic knowledge distillation showed the most promising results. Data variety for training and additional re-training is a helpful tool to boost HSV image segmentation quality. However, when performing re-training, the phenomenon of catastrophic forgetting should be kept in mind, i.e., adaption to new data while forgetting already learned knowledge.

BackgroundDuring the Covid-19 pandemic, singing activities were restricted due to several super-spreading events that have been observed during rehearsals and vocal performances. However, it has not been clarified how the aerosol dispersion, which has been assumed to be the leading transmission factor, could be reduced by masks which are specially designed for singers.Material and Methods12 professional singers (10 of the Bavarian Radio-Chorus and two freelancers, 7 females and 5 males) were asked to sing the melody of the ode of joy of Beethoven’s 9thsymphony “Freude schöner Götterfunken, Tochter aus Elysium” in D-major without masks and afterwards with five different singers’ masks, all distinctive in their material and proportions. Every task was conducted after inhaling the basic liquid from an e-cigarette. The aerosol dispersion was recorded by three high-definition video cameras during and after the task. The cloud was segmented and the dispersion was analyzed for all three spatial dimensions. Further, the subjects were asked to rate the practicability of wearing the tested masks during singing activities using a questionnaire.ResultsConcerning the median distances of dispersion, all masks were able to decrease the impulse dispersion of the aerosols to the front. In contrast, the dispersion to the sides and to the top was increased. The evaluation revealed that most of the subjects would reject performing a concert with any of the masks.ConclusionAlthough, the results exhibit that the tested masks could be able to reduce the radius of aerosol expulsion for virus-laden aerosol particles, there are more improvements necessary to enable the practical implementations for professional singing.

Purpose The pitch-shift reflex (PSR) is the adaptation of the fundamental frequency during phonation and speech and describes the auditory feedback control. Speakers without voice and speech disorders mostly show a compensation of the pitch change in the auditory feedback and adapt their fundamental frequency to the opposite direction. Dysphonic patients often display problems with the auditory perception and control of their voice during therapy. Our study focuses on the auditory and kinesthetic control mechanisms of patients with muscle tension dysphonia (MTD) and speakers without voice and speech problems. Main purpose of the study is the analysis of the functionality of the control mechanisms within phonation and speech between patients with MTD and normal speakers. Method Sixty-one healthy subjects (17 male, 44 female) and 22 patients with MTD (7 male, 15 female) participated following two paradigms including a sustained phonation (vowel /a/) and speech ([‘mama]). Within both paradigms the fundamental frequency of the auditory feedback was increased synthetically. For the analysis of the PSR the electroencephalogram, electroglottography, the voice signal, and the high-speed endoscopy data were recorded simultaneously. The PSR in the electroencephalogram was detected via the N100 and the mismatch negativity. Statistical tests were applied for the detection of the PSR in the physiological response within the electroglottography, voice, and high-speed endoscopy signals. The results were compared between both groups. Results No differences were found between the controls and patients with MTD regarding latency and magnitude of the perception of the pitch shift in both paradigms, but for the magnitude of the behavioral response. Differences also could be found for both groups between the “no pitch” and “pitch” condition of the two paradigms regarding vocal fold dynamics and voice quality. Patients with MTD showed more vibrational irregularities during the PSR than the controls, especially regarding the symmetry of vocal fold dynamics. Conclusion Patients with MTD seem to have a disturbed interaction between the auditory and kinesthetic feedback inducing the execution of an overriding behavioral response.

Surfing is becoming increasingly popular, and the economic market for surfing products is growing. The performance of the surfboards and the fins has evolved continuously through the experience and innovative spirit of surfers and shapers. However, provided performance characteristics are only subjective and of descriptive nature. Therefore, we apply and analyze objective performance indicators as (1) lift and drag coefficient for each fin and the entire configuration and (2) lift-to-drag ratio for the entire fin configuration. This numerical study investigated a commercial 3-fin configuration, mounted into a rectangular simulation region, considering the flow at several different angles of attack between 0°and 45°. RANS and URANS simulations were performed with the SST k − ω turbulence model at an inflow velocity of 5 m/s. Simulation results showed that the fins have an optimal range of attack angles where they produce the most lift. The lift forces generate a turning moment which forces the board to be turned in direction to the shore favoring the controllability and stability of the surfboard. The higher the angle of attack, the higher is the drag coefficient. Lift-to-drag ratio values showed that for speeding up small attack angles with low drag forces are preferable. Furthermore, unsteady effects as flow separation combined with vortex shedding occur at high angles of attack above 20°which can only be resolved by URANS simulations. These unsteady effects have high negative influences on stability due to high fluctuation amplitudes for lift and drag forces. In summary, this study presents, to the best of the authors’ knowledge, for the first time, steady and unsteady forces on surfboard fins and discusses their potential influences on the surfer’s controllability and stability of the surfboard during typical surfing maneuvers.

Ventricular folds (VeFs) act as passive, non-moving structures during normal phonation. According to the literature, VeFs potentially aid the flow-driven oscillations of the vocal folds (VFs) that produce the primary sound of human phonation. In this study, large eddy simulations were performed to analyze this influence in a numerical model with imposed VF motion as measured experimentally from a synthetic silicone vocal fold model. Model configurations with and without VeFs were considered. Furthermore, configurations with rectangular and elliptical glottis shapes were simulated to investigate the effects of three-dimensional glottal jet evolutions. Results showed that VeFs increased flow rate and transglottal pressure difference by a decrease in the pressure level in the ventricles immediately downstream of the VFs. This led to an increase in the glottal flow resistance, increased energy transfer rate between the flow and VFs, and a simultaneous decrease in the laryngeal flow resistance, which shows a higher amount of kinetic energy in the glottal flow. This enhancement was more pronounced in the rectangular glottis and varied with the subglottal pressure and VeF gap size.

The U-Net architecture [1] is a state-of-the-art neural network for semantic image segmentation that is widely used in biomedical research. It is based on an encoder-decoder framework and its vanilla version shows already high performance in terms of segmentation quality. Due to its large parameter space, however, it has high computational costs on both, CPUs and GPUs. In a research setting, inference time is relevant, but not crucial for the results.

Summary Introduction Vocal registers and the frequency region where registration events occur, the passaggio, have been in focus of scientific research for almost 200 years. In professional tenors, it has been shown before that singing across the passaggio avoiding a register shift and therefore using their stage voice above the passaggio (SVaP) is associated with greater vocal stability than a register change to the falsetto. However, it is unclarified how much different loudness conditions contribute to this vocal stability. Material and methods Six professional tenors were asked to perform four pitch glides from A3 to A4 (220-440 Hz) on the vowel [i:]. These glides included (1) the passaggio from modal register to falsetto. The following glides into SVaP were performed under different loudness conditions, (2) mezzoforte (average loudness), (3) pianissimo (as quietly as possible), and (4) fortissimo (the loudest possible). During phonation, high speed videoendoscopy (HSV), electroglottography, and audio signals were recorded simultaneously. The glottal area waveform was derived based on the HSV material. Results Modal to falsetto transitions were associated with relatively low sound pressure level and rise of open quotients (OQ) for the falsetto. Transitions to SVaP showed a clear dependence on the intended loudness. The OQs were lower the louder the task was. There was no clear evidence that transitions with softer voice showed greater stability of vocal fold oscillation patterns than louder tasks. Conclusions The vocal fold oscillation pattern show- differences among various loudness conditions within the tenors’ passaggio but no clear differences with regard to oscillatory stability.

Assistant physicians typically learn surgical techniques by observation and supervised practice on the patient or using biophantoms. Alternatively, surgical simulators with the possibility of new training possibilities can be used. A number of simulators is already commercially available and might in the future become as important for surgical training as flight simulators. Since so far no simulator is concerned with the training of tracheotomies, a soft-tissue model for simulating tracheotomy was developed. This soft-tissue model is integrated into our ENT surgical simulator for tracheotomy. To model the soft-tissues of the neck (skin and fat), a computed tomography (CT) scan was interactively segmented. For the interaction simulation of a scalpel with the soft-tissue, position based dynamics (PBD) was used, originally developed for the gaming industries. Initial results imply that the proposed approach is able to model soft tissues for virtual surgical training.

Wearing face coverings became one essential tool in order to prohibit virus transmission during the COVID-19 pandemic. In comparison to speaking and breathing, singing emits a much higher amount of aerosol particles. Therefore, there are situations in which singers can perform or rehearse only if they are using protective masks. However, such masks have a more or less adverse effect not only on the singer’s comfort and tightness of the mask but also on the radiated sound. For this reason, the spectral filtering and directivity of masks designed specifically for professional singing was measured. The tests were performed with a head phantom. Over most of the spectrum, attenuation is observed, although amplification happens at some low frequency bands for different mask types and directions. Especially singing masks with a plastic face shield showed partial amplification of up to +10 dB below a frequency of 2 kHz, while only slight significant attenuation and no amplification (minimal acoustic loss) were seen for woven fabric masks. Above 2.5 kHz, the transparent masks showed the greatest sound attenuation up to −30 dB, while woven fabric masks produced an overall lower sound attenuation of up to −5 dB. In addition at low frequencies, the sound was amplified or attenuated equally in all directions for masks with a stiff plastic face shield. At higher frequencies, the attenuation is higher to the frontal than to the backward direction.

Summary Objective The aim of the study was to establish a basic data set of combined functional and anatomical measures of aged sheep larynges using ex vivo models. Combining these two approaches in one and the same larynx is an unmet goal so far yet is important as newer treatment strategies aim to preserve the organ structure and new assessment tools are required. Ovine larynges were used as their dimensions, and muscle fiber type distribution highly resemble the human larynx. Study Design Ex vivo animal study. Methods Larynges of six sheep (~9 years of age) were subjected to ex vivo functional phonatory experiments. Phonatory characteristics were analyzed as a function of longitudinal vocal fold (VF) prestress. Anatomical measurements of the same larynges comprised micro-computed tomography scans followed by three-dimensional (3D) reconstructions. Using specially adapted radiological scan protocols with subsequent 3D reconstruction, muscle volumes, surface areas, and anatomical measurements were computed. Results Increasing longitudinal prestress yielded higher subglottal pressure (P S ) for the same airflow. Quantitative differences to previous studies—such as the increased P S and increased phonation threshold pressure—were detected. We achieved excellent visualization of the laryngeal muscles and framework, resulting in accurate 3D reconstructions for quantitative analysis. We found no significant intraindividual volume differences of the thyroarytenoid muscles. Conclusion The established protocol allows precise functional and anatomical measures. The data created provide a reference data set for upcoming therapeutic strategies (eg, growth factor therapy, functional electrical stimulation) that target essential structures of the VFs such as the laryngeal muscles and/or the VF mucosa.

OBJECTIVE/HYPOTHESIS Laser-based three-dimensional (3D) imaging allows realistic visualization and absolute measurement of the vocal folds for comprehensive characterization of the oscillation pattern. STUDY DESIGN Exploratory study with descriptive data analysis in healthy cohort. METHODS The custom-developed miniaturized measuring device is an endoscopic camera-laser setup for stereo triangulation and suitable for in vivo application. The laser projection unit generates a regular laser grid that is projected on the vocal folds and recorded at 4000 fps using a high-speed camera. Recordings are performed during sustained phonation on 10 healthy subjects. RESULTS We present absolute values for the lateral and vertical displacement amplitudes and maximum velocities during opening and closing phase. On average over all study participants, the vertical parameters are at least 50% higher than their lateral counterparts. The mean male/female amplitudes are 0.93 mm/0.80 mm in the lateral and 1.28 mm/1.45 mm in the vertical direction. Only 20% of the healthy subjects displayed slight asymmetries in the lateral direction, but 70% showed significant asymmetries in the vertical component. In only 30% of all subjects, the asymmetry trends matched in the lateral and vertical direction. CONCLUSIONS The study illustrates the additional benefit of 3D imaging: consideration of the vertical component and computation of metric units. New, relevant, and even unexpected information was obtained by the incorporation of the additional vertical dimension. Metric units allow for an absolute evaluation of anatomy and dynamics and therefore enables cross-study comparability. The noninvasive acquisition of quantitative measures for evidence-based medicine has the potential to enhance diagnostic and therapeutic procedures as well as basic medical research. LEVEL OF EVIDENCE 4. Laryngoscope, 128:2367-2374, 2018.

This study presents a framework for a direct comparison of experimental vocal fold dynamics data to a numerical two-mass-model (2MM) by solving the corresponding inverse problem of which parameters lead to similar model behavior. The introduced 2MM features improvements such as a variable stiffness and a modified collision force. A set of physiologically sensible degrees of freedom is presented, and three optimization algorithms are compared on synthetic vocal fold trajectories. Finally, a total of 288 high-speed video recordings of six excised porcine larynges were optimized to validate the proposed framework. Particular focus lay on the subglottal pressure, as the experimental subglottal pressure is directly comparable to the model subglottal pressure. Fundamental frequency, amplitude and objective function values were also investigated. The employed 2MM is able to replicate the behavior of the porcine vocal folds very well. The model trajectories’ fundamental frequency matches the one of the experimental trajectories in $$98.6\%$$ of the recordings. The relative error of the model trajectory amplitudes is on average $$9.5\%$$ . The experiments feature a mean subglottal pressure of 10.16 (SD $$= 2.31$$ ) $${\text {cmH}}_2{\text {O}}$$ ; in the model, it was on average 7.61 (SD $$= 2.40$$ ) $${\text {cmH}}_2{\text {O}}$$ . A tendency of the model to underestimate the subglottal pressure is found, but the model is capable of inferring trends in the subglottal pressure. The average absolute error between the subglottal pressure in the model and the experiment is 2.90 (SD $$= 1.80$$ ) $${\text {cmH}}_2{\text {O}}$$ or $$27.5\%$$ . A detailed analysis of the factors affecting the accuracy in matching the subglottal pressure is presented.

Ex vivo larynx experiments are limited in time due to degeneration of the laryngeal tissues. In order to acquire a significant and comparable amount of data, automatization of current manual experimental procedures is desirable. A computer controlled, electro-mechanical setup was developed for time-dependent variation of specific physiological parameters, including adduction and elongation level of the vocal folds and glottal flow. The setup offers a standardized method to induce defined forces on the laryngeal cartilages. Furthermore, phonation onset is detected automatically and the subsequent measurement procedure is automated and standardized to improve the efficiency of the experimental process. The setup was validated using four ex vivo porcine larynges, whereas each validation measurement series was executed with one separate larynx. Altogether 31 single measurements were undertaken, which can be summed up to a total experimental time of about 4 min. Vocal fold elongation and adduction lead both to an increase in fundamental frequency and subglottal pressure. Measurement procedures like applying defined subglottal pressure steps and onset-offset detection were reliably executed. The setup allows for a computer-based parameter control, which enables fast experimental execution over a wide range of laryngeal configurations. This maximizes the number of measurements and reduces personal effort compared with manual procedures.

OBJECTIVES/HYPOTHESIS High-speed videoendoscopy (HSV) has potential to objectively quantify vibratory vocal fold characteristics during phonation. Glottal Analysis Tools (GAT) version 2018, developed in Erlangen, Germany, is software for determining various glottal area waveform (GAW) quantities. Before having GAT analyze HSV videos, segmenters have to define glottis manually across videos in a semiautomatic segmentation protocol. Such interventions are hypothesized to induce variability of subsequent GAW measure computation across segmenters and may attenuate GAT measures' reliability to a certain point. This study explored intersegmenter variability in GAT's GAW measures based on semiautomatic image processing. STUDY DESIGN Cohort study of rater reliability. METHODS In total, 20 HSV videos from normophonic and dysphonic subjects with various laryngeal disorders were selected for this study and segmented by three trained segmenters. They separately segmented glottis areas in the same frame sets of the videos. Upon analysis of GAW, GAT offers 46 measures related to topologic GAW dynamic characteristics, GAW periodicity and perturbation characteristics, and GAW harmonic components. To address GAT's reliability, intersegmenter-based variability in these measures was examined with intraclass correlation coefficient (ICC). RESULTS In general, ICC behavior of the 46 GAW measures across three raters was highly acceptable. ICC of one parameter was moderate (0.5

Objectives/hypothesis To evaluate the demographics, clinical features, management, and prognostic indicators of tracheoesophageal puncture complications in patients undergoing placement of voice prosthesis following cancer treatment. Study design Retrospective analysis. Methods A retrospective analysis was conducted of cases from a tertiary referral center diagnosed between 1996 and 2015. Multivariate logistic regression was used to determine factors associated with tracheoesophageal puncture (TEP) and voice prostheses-complication-free survival (TEP/VP-CFS). Results One hundred fourteen cases were identified. Most patients were males (92.9%) with pT3 (26.8%) or pT4 (58.1%) N+ (53.6%) tumors. All patients received laryngectomy as the primary treatment, with 75% of patients receiving adjuvant radiation therapy or chemoradiotherapy. Complications with TEP were common (65.2%). The most frequent problem was salivary leakage (50.0%), which at the same time was the most common reason for changing the prosthesis. On univariate regression analysis, prosthesis placement time after adjuvant radiotherapy (hazard ratio [HR]: 4.17, 95% confidence interval [CI]: 2-8.69), secondary prosthesis placement after primary surgery (HR: 3.97, 95% CI: 1.99-7.9), and laryngectomy with flap reconstruction (HR: 1.96, 95% CI: 0.99-3.89) were significant prognosticators for complications. Multivariate regression analysis revealed secondary prosthesis placement after adjuvant radiotherapy (HR: 3.66, 95% CI: 1.39-9.68) or after primary surgery (HR: 2.57, 95% CI: 0.92-7.2) to be the strongest predictors of reduced TEP/VP-CFS. Conclusions Secondary prosthesis placement after primary surgery, placement after previous irradiation, and laryngectomy with flap reconstruction are predictors of poor TEP/VP-CFS. Planned adjuvant radiotherapy is not a contraindication for TEP with prosthetic placement, but it is very important to place the prosthesis during the primary surgery or at least before scheduled radiotherapy. Level of evidence 4 Laryngoscope, 2020.

Summary Introduction It is generally assumed that when singing across the region where registration events for untrained voices occur (the passaggio), singers modify the voice production system in order to avoid changes of voice quality. In this context, it has been postulated that nasalance could be used to stabilize vocal function throughout the passaggio. However, whether nasalance is frequently used by professional singers and if so, if it has a stabilizing effect on vocal fold oscillation patterns, is not yet fully understood. Material and Methods Eight western classically trained professional male singers (6 tenors and 2 baritones) were asked to perform transitions (1) from modal to falsetto register and (2) from modal to stage voice above the passaggio (SVaP) during ascending pitch glides from A3 (ƒo approx. 220 Hz) to A4 (ƒo approx. 440 Hz) on the vowel /i/. Transnasal high-speed videoendoscopy at 20.000 fps was captured simultaneously with electroglottographic, nasal and oral flow, and audio signals, recorded using the same frame rate. The nasalance was calculated from both oral and nasal DC-flow signals. Results Transitions to SVaP showed greater periodicity and regularity than transitions to falsetto. For 5 subjects, nasalance was increased during the passaggio for the transition to SVaP. For 4 subjects the increase of nasalance for the SVaP was associated with a stabilization of the open quotient and occurred at a comparable fundamental frequency as the increase of the open quotient for the transition to falsetto. Conclusions Nasalance can be used in order to stabilize oscillatory regularity and open quotient in male singers for singing across the passaggio.

Objectives/Hypothesis In professional classical singing, an even voice quality throughout the entire singing voice range is essential. Transitions between vocal registers (passaggio) are the technically most challenging aspects in classical singing. It is hypothesized that they are most affected by vocal fold mass lesions (VFML). Study Design Cohort study. Methods In this study, the effect of VFML on vocal fold vibration in the passaggio regions was analyzed in four female and three male singers suffering from organic dysphonia. The singers were asked to sing an ascending glissando through the passaggio regions, before and after treatment. The vocal fold vibration was documented with transnasal endoscopic high-speed imaging recordings at 20,000 frames per second, supplemented by synchronized acoustic and electroglottographic recordings. Results Major irregularities were found in the passaggio region of four singers before treatment, whereas the respective phonations below the passaggio were almost regular. In two female singers only the upper, but not the lower passaggio was affected. In all four of these participants, the passaggio region was more regular after treatment. In the remaining three participants, the VFML showed no effect on the passaggio region. However, the singers' ability to reach higher pitches was impaired, but was resolved after treatment. Conclusions The data in this case study strongly suggest that the passaggio region could be affected by VFML, even if phonation outside the passaggio regions is unimpaired. When planning surgical procedures for professional singers, clinical examination protocols should therefore include phonatory tests across the passaggio regions. Level of Evidence 4 Laryngoscope, 127:1392–1401, 2017

Ex-vivo hemilarynx experiments allow the visualization and quantification of three-dimensional dynamics of the medial vocal fold surface. For three excised human male larynges, the vibrational output, the glottal flow resistance, and the sound pressure during sustained phonation were analyzed as a function of vocal fold adduction for varying subglottal pressure. Empirical eigenfunctions, displacements, and velocities were investigated along the vocal fold surface. For two larynges, an increase of adduction level resulted in an increase of the glottal flow resistance at equal subglottal pressures. This caused an increase of lateral and vertical oscillation amplitudes and velocity indicating an improved energy transfer from the airflow to the vocal folds. In contrast, the third larynx exhibited an amplitude decrease for rising adduction accompanying reduction of the flow resistance. By evaluating the empirical eigenfunctions, this reduced flow resistance was assigned to an unbalanced oscillation pattern with predominantly lateral amplitudes. The results suggest that adduction facilitates the phonatory process by increasing the glottal flow resistance and enhancing the vibrational amplitudes. However, this interrelation only holds for a maintained balanced ratio between vertical and lateral displacements. Indeed, a balanced vertical-lateral oscillation pattern may be more beneficial to phonation than strong periodicity with predominantly lateral vibrations.

Heutige visuelle Untersuchungsmethoden des Kehlkopfes und der Stimmlippenbewegung erlauben vorwiegend eine zweidimensionale Darstellung der real dreidimensionalen Strukturen und Bewegungen. Die vertikale Bewegungskomponente der Stimmlippen wird meist vernachlassigt, was zu einem Informationsverlust fuhrt. Neuere Studien zeigen, dass die vertikale Komponente in einer ahnlichen Grosenordnung wie die laterale Komponente liegt und einen nicht unerheblichen Anteil am Phonationsprozess hat. In der vorliegenden Arbeit wird ein Verfahren zur zukunftigen 3D-Rekonstruktion und Visualisierung endoskopisch aufgezeichneter Stimmlippenschwingungen prasentiert. Das Setup beinhaltet eine Hochgeschwindigkeitskamera (HGK) und ein Laserprojektionssystem (LPS) welches ein regelmasiges Punktegitter auf die Stimmlippen projiziert. Das System ermoglicht die 3D-Rekonstruktion der Stimmlippenoberflachen und quantitative Aussage uber dreidimensionale Auslenkungseigenschaften und Geschwindigkeitswerte. Die Anwendbarkeit der Methodik wird jeweils an einem humanen Kadaverkehlkopf, einem Schweinekehlkopf sowie an einem kunstlichen Kehlkopfmodell mit Silikonstimmlippen demonstriert. Der Versuchsaufbau ermoglicht die Rekonstruktion der Topographie der sichtbaren Oberflache beider Stimmlippen zu jedem Aufnahmezeitpunkt der HGK, wodurch sich die dreidimensionale Oberflachenbewegung der Stimmlippen wahrend der Phonation berechnen lasst. Das Setup erlaubt die quantitative Analyse der Stimmlippenphysiologie bzgl. Auslenkungen, Geschwindigkeiten und Beschleunigungseigenschaften. Das nachste Ziel ist die Miniaturisierung des LPS fur klinische Untersuchungen am Menschen. Hier erwarten wir uns neue Erkenntnisse uber die Zusammenhange zwischen 3D-Schwingungseigenschaften und akustischer Signalqualitat bei gesunder und kranker Stimmgebung.

Die Stimmgebung ist ein Prozess, der noch immer in vielen Details nicht verstanden ist. Neben der beobachtenden Forschung versuchen seit langerer Zeit Grundlagenwissenschaftler die Physik der Stimme zu erklaren, indem sie versuchen, die Stimmbildung nachzubauen. Dieser Prozess wird als Modellierung bezeichnet. Die Erkenntnisse dieses Forschungszweigs sind vielen Berufsgruppen, die mit Stimme zu tun haben, wie Gesangspadagogen, Stimmtherapeuten und Sprecherziehern, aber auch zum Teil HNO-Arzten und Phoniatern, nicht immer gelaufig. Der Grund liegt in der Tatsache, dass sie zum Teil mathematisch und physikalisch sehr komplexe Beschreibungen enthalten, die nicht immer schnell erfasst werden konnen. Vereinfachte Darstellung und Erklarung aktueller Modelle zur Modellierung des Phonationsprozesses, die in den letzten Jahren viel dazu beigetragen haben, die Zusammenhange bei der Stimmgebung besser zu verstehen. Die vorgestellten Modelle decken ein breites Gebiet von numerisch relativ einfachen bis mathematisch hoch komplexen Modellen ab. Experimentelle Modelle basieren z. B. auf schwingenden Silikonstimmlippen oder sind nur statischer Natur. Kadaverkehlkopfmodelle zeichnen sich durch ihre Realitatsnahe aus. Es wird ein Uberblick uber Modellarten der Stimmgebungssimulation gegeben, um aufzuzeigen, welche Methoden zur Simulation der Stimmgebung existieren, ohne aber auf die komplexen zu Grunde liegenden mathematischen oder physikalischen Sachverhalte im Detail einzugehen. Die numerische oder experimentelle Modellierung des Stimmgebungsprozesses ermoglicht die Aufdeckung von Kausalitaten, die zukunftig genutzt werden konnen, um sowohl konservative als auch operative Behandlungsmethoden zu adaptieren oder neue Methoden zu entwickeln.

The current use of laryngeal high-speed videoendoscopy in clinic settings involves subjective visual assessment of vocal fold vibratory characteristics. However, objective quantification of vocal fold vibrations for evidence-based diagnosis and therapy is desired, and objective parameters assessing laryngeal dynamics have therefore been suggested. This study investigated the sensitivity of the objective parameters and their dependence on recording frame rate. A total of 300 endoscopic high-speed videos with recording frame rates between 1000 and 15 000 fps were analyzed for a vocally healthy female subject during sustained phonation. Twenty parameters, representing laryngeal dynamics, were computed. Four different parameter characteristics were found: parameters showing no change with increasing frame rate; parameters changing up to a certain frame rate, but then remaining constant; parameters remaining constant within a particular range of recording frame rates; and parameters changing with nearly every frame rate. The results suggest that (1) parameter values are influenced by recording frame rates and different parameters have varying sensitivities to recording frame rate; (2) normative values should be determined based on recording frame rates; and (3) the typically used recording frame rate of 4000 fps seems to be too low to distinguish accurately certain characteristics of the human phonation process in detail.

The human voice originates from oscillations of the vocal folds in the larynx. The duration of the voice onset (VO), called the voice onset time (VOT), is currently under investigation as a clinical indicator for correct laryngeal functionality. Different analytical approaches for computing the VOT based on endoscopic imaging were compared to determine the most reliable method to quantify automatically the transient vocal fold oscillations during VO. Transnasal endoscopic imaging in combination with a high-speed camera (8000 fps) was applied to visualize the phonation onset process. Two different definitions of VO interval were investigated. Six analytical functions were tested that approximate the envelope of the filtered or unfiltered glottal area waveform (GAW) during phonation onset. A total of 126 recordings from nine healthy males and 210 recordings from 15 healthy females were evaluated. Three criteria were analyzed to determine the most appropriate computation approach: (1) reliability of the fit function for a correct approximation of VO; (2) consistency represented by the standard deviation of VOT; and (3) accuracy of the approximation of VO. The results suggest the computation of VOT by a fourth-order polynomial approximation in the interval between 32.2 and 67.8% of the saturation amplitude of the filtered GAW.

Vocal fold kinematics and its interaction with aerodynamic characteristics play a primary role in acoustic sound production of the human voice. Investigating the temporal details of these kinematics using high-speed videoendoscopic imaging techniques has proven challenging in part due to the limitations of quantifying complex vocal fold vibratory behavior using only two spatial dimensions. Thus, we propose an optical method of reconstructing the superior vocal fold surface in three spatial dimensions using a high-speed video camera and laser projection system. Using stereo-triangulation principles, we extend the camera-laser projector method and present an efficient image processing workflow to generate the three-dimensional vocal fold surfaces during phonation captured at 4000 frames per second. Initial results are provided for airflow-driven vibration of an ex vivo vocal fold model in which at least 75% of visible laser points contributed to the reconstructed surface. The method captures the vertical motion of the vocal folds at a high accuracy to allow for the computation of three-dimensional mucosal wave features such as vibratory amplitude, velocity, and asymmetry.

Most sports like surfing are highly developed. It is necessary to tease the last percentages out of the competitors and equipment&mdash, in the case of surfing the surfboard-fin-system&mdash, to win competitions or championships. In this computational investigation, a parameter study of the positioning of the two rear fins within a 4-fin configuration with fixed front fins on a surfboard is executed to find appropriate fin positions for specific surf situations. Four different inflow velocities are investigated. The RANS and URANS models combined with the SST k-&omega, turbulence model, which is available within the computational fluid dynamics (CFD) package STAR-CCM+, are used to simulate the flow field around the fins for angles of attack (AoA) between 0&deg, and 45&deg, The simulation results show that shifting the rear fins toward the longitudinal axis of the surfboard lowers the maximum lift. Surfboards with 4-fin configurations are slower in nearly the whole range of AoA due to a higher drag force but produce a higher lift force compared to the 3-fin configuration. The lift and drag forces increase significantly with increasing inflow velocity. This study shows a significant influence of the rear fin positioning and the inflow velocity on lift and drag performance characteristics.

Laryngeal endoscopy is one of the primary diagnostic tools for laryngeal disorders. The main techniques are videostroboscopy and lately high-speed video endoscopy. Unfortunately, due to the restricting anatomy of the larynx and technical limitations of the recording equipment, many videos suffer from insufficient illumination, which complicates clinical examination and analysis. This work presents an approach to enhance low-light images from high-speed video endoscopy using a convolutional neural network. We introduce a new technique to generate realistically darkened training samples using Perlin noise. Extensive data augmentation is employed to cope with the limited training data allowing training with just 55 videos. The approach is compared against four state-of-the-art low-light enhancement methods and statistically significantly outperforms each on a no-reference (NIQE) and two full-reference (PSNR, SSIM) image quality metrics. The presented approach can be run on consumer-grade hardware and is thereby directly applicable in a clinical context. It is likely transferable to similar techniques such as videostroboscopy. Graphical Abstract The basic setup for training and employing an improved fully convolutional U-Net neural network to predict a brightness map used to enhance the lighting of ill-lit endoscopic high-speed videos - Artificially darkened training data are created using Perlin noise to allow region-specific darkening.