# Just Because We Camp, Doesn't Mean We Should: The Ethics of Modelling Queer Voices

arXiv: https://arxiv.org/abs/2406.07504 · HTML: https://mlsanity.com/paper/2406.07504/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 8 of its 32 references that are in the map.
- First posted on arXiv 11 Jun 2024.

## Builds on (most cited first, 8 of 8)

- [Neural Codec Language Models are Zero-Shot Text to Speech Synthesizers](https://mlsanity.com/paper/2301.02111/) (2023)
- [TED-LIUM 3: twice as much data and corpus repartition for experiments on speaker adaptation](https://mlsanity.com/paper/1805.04699/) (2018)
- [Attention is All You Need in Speech Separation](https://mlsanity.com/paper/2010.13154/) (2020)
- [The Values Encoded in Machine Learning Research](https://mlsanity.com/paper/2106.15590/) (2021)
- [ICASSP 2022 Deep Noise Suppression Challenge](https://mlsanity.com/paper/2202.13288/) (2022)
- [Stereotypes and Smut: The (Mis)representation of Non-cisgender Identities by Text-to-Image Models](https://mlsanity.com/paper/2305.17072/) (2023)
- [Ethics Sheets for AI Tasks](https://mlsanity.com/paper/2107.01183/) (2021)
- [Discipline and Label: A WEIRD Genealogy and Social Theory of Data Annotation](https://mlsanity.com/paper/2402.06811/) (2024)

## Built on by (most cited first, 3 of 3)

- [Sound Check: Auditing Audio Datasets](https://mlsanity.com/paper/2410.13114/) (2024)
- [Ethics Whitepaper: Whitepaper on Ethical Research into Large Language Models](https://mlsanity.com/paper/2410.19812/) (2024)
- [Queer inclusion in speech datasets: An audit and taxonomy of practical tensions](https://mlsanity.com/paper/2609.25491/) (2026)

## Abstract (arXiv)

Modern voice cloning models claim to be able to capture a diverse range of voices. We test the ability of a typical pipeline to capture the style known colloquially as "gay voice" and notice a homogenisation effect: synthesised speech is rated as sounding significantly "less gay" (by LGBTQ+ participants) than its corresponding ground-truth for speakers with "gay voice", but ratings actually increase for control speakers. Loss of "gay voice" has implications for accessibility. We also find that for speakers with "gay voice", loss of "gay voice" corresponds to lower similarity ratings. However, we caution that improving the ability of such models to synthesise ``gay voice'' comes with a great number of risks. We use this pipeline as a starting point for a discussion on the ethics of modelling queer voices more broadly. Collecting "clean" queer data has safety and fairness ramifications, and the resulting technology may cause harms from mockery to death.