Elizabeth Erickson‐DiRenzo

Objectives Advances in artificial intelligence (AI) technology have increased the feasibility of classifying voice disorders using voice recordings as a screening tool. This work develops upon previous models that take in single vowel recordings by analyzing multiple vowel recordings simultaneously to enhance prediction of vocal pathology. Methods Voice samples from the Saarbruecken Voice Database, including three sustained vowels (/a/, /i/, /u/) from 687 healthy human participants and 334 dysphonic patients, were used to train 1‐dimensional convolutional neural network models for multiclass classification of healthy, hyperfunctional dysphonia, and laryngitis voice recordings. Three models were trained: (1) a baseline model that analyzed individual vowels in isolation, (2) a stacked vowel model that analyzed three vowels (/a/, /i/, /u/) in the neutral pitch simultaneously, and (3) a stacked pitch model that analyzed the /a/ vowel in three pitches (low, neutral, and high) simultaneously. Results For multiclass classification of healthy, hyperfunctional dysphonia, and laryngitis voice recordings, the stacked vowel model demonstrated higher performance compared with the baseline and stacked pitch models (F1 score 0.81 vs. 0.77 and 0.78, respectively). Specifically, the stacked vowel model achieved higher performance for class‐specific classification of hyperfunctional dysphonia voice samples compared with the baseline and stacked pitch models (F1 score 0.56 vs. 0.49 and 0.50, respectively). Conclusions This study demonstrates the feasibility and potential of analyzing multiple sustained vowel recordings simultaneously to improve AI‐driven screening and classification of vocal pathology. The stacked vowel model architecture in particular offers promise to enhance such an approach. Lay Summary AI analysis of multiple vowel recordings can improve classification of voice pathologies compared with models using a single sustained vowel and offer a strategy to enhance AI‐driven screening of voice disorders. Level of Evidence 3

Objective Transfeminine patients (transwomen/feminine nonbinary folks assigned male at birth) can undergo chondrolaryngoplasty (“tracheal shave”) to feminize their neck appearance. While isolated cases of vocal complications have been reported following the procedure, aggregated outcomes have not been quantitatively studied. We present acoustic and stroboscopic data to describe a patient cohort with vocal complications after chondrolaryngoplasty and discuss reparative surgical technique. Methods Subjective and objective data, including videostroboscopy, were collected from patients with voice complaints after chondrolaryngoplasty. Dislocated anterior commissures were reconstructed with feminization laryngoplasty. Postoperative voice data were recorded and statistically compared to preoperative data using paired t ‐tests. Results On consecutive chart review, of the 94 transfeminine women with prior outside history of chondrolaryngoplasty, 27 (29%) reported chronic postoperative hoarseness, deepened pitch, or loss of upper register. On endoscopy, short, lax vocal folds with persistent anterior glottic gap and phase asymmetry were commonly noted; anterior commissure dislocation was confirmed in‐office by using needle localization through absent thyroid cartilage. After open resuspension of the anterior commissure with feminization laryngoplasty, post‐repair modal‐speaking, minimum, and maximum fundamental frequencies (F0) increased on average by 7, 8, and 5 semitones, respectively ( p < 0.01), when compared to pre‐repair values. On average, perioperative maximum phonation time did not change significantly ( p = 0.15). Average self‐assessment of vocal femininity increased by 48% ( p < 0.01). Conclusion Anterior commissure dislocation should be suspected with signs of vocal impairment after chondrolaryngoplasty. Following proper diagnosis, resuspension of the anterior commissure via feminization laryngoplasty approach can be an effective reparative technique. Level of Evidence This work represents a 2011 OCEBM Level 4 evidence as a case series Laryngoscope , 133:2301–2307, 2023

Objective(s) To quantify the effect of laryngeal prominence size on socially perceived attributes relating to gender expression. Chondrolaryngoplasty (“tracheal shave”) is a common procedure performed for transgender women to feminize neck appearance. The extent of thyroid cartilage resection needed to convey socially‐perceived feminine gender expression without destabilizing the voice is incompletely understood. Methods Cross‐sectional evaluation of a randomized allocation of images of varying laryngeal prominence to a non‐repeated, random sample from November 2021 to December 2021. Photos of laryngeal prominence were isolated against a constant neck baseline with lateral, oblique, and frontal views. The images were embedded into a web‐based survey with visual analog scales to capture perceived scaled gender expression (masculinity, femininity) and social traits (e.g., attractiveness, friendliness, leadership). We performed bivariate and multivariate analyses relating the laryngeal prominence to perceived gender expression and social traits. Results The analytic sample included 1,026 respondents. Laryngeal grades similar to the demonstrated “grade M” in this study and smaller demonstrated similar perceptions of increased femininity and decreased masculinity. Grades larger than M demonstrate significantly increased perceived masculinity and significantly decreased perceived femininity. The lateral and oblique views of the neck appear to be the most gender‐informative. Conclusion This crowd‐sourced analysis of external laryngeal anatomy by a large population of observers provides clear, reproducible insights into social perceptions of gender identity and specifically femininity. These data will meaningfully inform patient counseling and surgical planning for gender‐affirming interventions by establishing normative data representing the general public's perceptions. Level of Evidence NA Laryngoscope , 133:2292–2300, 2023