Speech After Gender: A Trans-Feminine Perspective on Next Steps for Speech Science and Technology
This addresses limitations in speech technology for trans-feminine individuals, highlighting incremental improvements in speaker modeling.
The paper tackles the problem that current speaker modeling approaches fail to account for voice flexibility in gender modification, showing that gender classification systems are highly sensitive to voice changes and speaker verification systems fail with drastic modifications. They propose modeling individual vocal qualities like pitch and resonance as a solution.
As experts in voice modification, trans-feminine gender-affirming voice teachers have unique perspectives on voice that confound current understandings of speaker identity. To demonstrate this, we present the Versatile Voice Dataset (VVD), a collection of three speakers modifying their voices along gendered axes. The VVD illustrates that current approaches in speaker modeling, based on categorical notions of gender and a static understanding of vocal texture, fail to account for the flexibility of the vocal tract. Utilizing publicly-available speaker embeddings, we demonstrate that gender classification systems are highly sensitive to voice modification, and speaker verification systems fail to identify voices as coming from the same speaker as voice modification becomes more drastic. As one path towards moving beyond categorical and static notions of speaker identity, we propose modeling individual qualities of vocal texture such as pitch, resonance, and weight.