Total: 1
Gender-affirming voice training (GAVT) can reduce voice-gender incongruence for transgender and gender-diverse individuals, but access is often limited by high costs and a lack of qualified providers. Interactive software could expand access, but existing GAVT apps are typically limited in functionality. This paper describes the development and testing of software to address one of the most challenging aspects of GAVT: modifying vocal tract resonance. We introduce a biofeedback system that uses a real-time linear predictive coding (LPC) spectrum to visualize changes in resonance as learners adjust their vocal tract shape. Visual targets for brighter (feminine) and darker (masculine) resonances help guide users toward their desired voice characteristics. In-lab user testing with 10 trans women yielded an average System Usability Scale (SUS) score of 75.25, supporting the acceptability of the tool as an adjunct resonance training in the GAVT context.