Long-Term Average Spectrum Analysis of Voice After Endoscopic Vocal Fold Shortening and Therapy

Endoscopic Vocal Fold Shortening (EVFS) is a common gender-affirming surgical procedure for transgender women, yet postoperative acoustic outcomes have largely been described using only fundamental frequency (f _o) and traditional clinical measures, leaving broader acoustic characterization limited. This study investigated the utility of Long-Term Average Spectrum (LTAS) analysis in evaluating spectral changes after EVFS, identifying spectral markers to better capture surgical and therapeutic effects on voice feminization.

Voice recordings were collected, and LTAS was analyzed at baseline and at 3 and 12 months postoperatively. Spectral slope, Formant Cluster Prominence FCP [dB], and FCP (Hz) were examined across timepoints. At 12-months, transgender participants were compared to cisgender male and female control voices for reference.

At 3 months, spectral energy increased in the higher frequency range, spectral slope flattened, and FCP [dB] was significantly reduced. By 12 months, spectral slope remained flatter than baseline, and FCP [dB] demonstrated a modest rebound, but remained below preoperative levels. Group comparisons demonstrated clear approximation of transgender women toward cisgender female spectral profiles at 12 months, with divergence from cisgender male patterns. LTAS parameters-including spectral slope and FCP [dB]-showed marked convergence with cisgender female characteristics. Importantly, sound pressure level (SPL) remained stable, suggesting spectral changes primarily reflected structural and behavioral adaptation rather than intensity.

LTAS provides valuable insight into the spectral dynamics of voice feminization after EVFS beyond f _o. Observed spectral changes over time suggest behavioral adaptation to surgical modification may extend for up to a year. Thus, LTAS emerges as a useful tool for monitoring recovery and adaptation.