# I Hear, Therefore I Trust: A Socio-Technical Investigation of Humans as Synthetic Speech Detectors

arXiv: https://arxiv.org/abs/2605.28064 · HTML: https://mlsanity.com/paper/2605.28064/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 13 of its 56 references that are in the map.
- First posted on arXiv 27 May 2026.

## Builds on (most cited first, 13 of 13)

- [LibriTTS: A Corpus Derived from LibriSpeech for Text-to-Speech](https://mlsanity.com/paper/1904.02882/) (2019)
- [YourTTS: Towards Zero-Shot Multi-Speaker TTS and Zero-Shot Voice Conversion for everyone](https://mlsanity.com/paper/2112.02418/) (2021)
- [ELEPHANT: Measuring and understanding social sycophancy in LLMs](https://mlsanity.com/paper/2505.13995/) (2025)
- [ASVspoof 2021: Towards Spoofed and Deepfake Speech Detection in the Wild](https://mlsanity.com/paper/2210.02437/) (2022)
- [Emotions Don't Lie: An Audio-Visual Deepfake Detection Method Using Affective Cues](https://mlsanity.com/paper/2003.06711/) (2020)
- [ASVspoof 5: Crowdsourced Speech Data, Deepfakes, and Adversarial Attacks at Scale](https://mlsanity.com/paper/2408.08739/) (2024)
- [Where are we in audio deepfake detection? A systematic analysis over generative and detection models](https://mlsanity.com/paper/2410.04324/) (2024)
- [Unmasking Illusions: Understanding Human Perception of Audiovisual Deepfakes](https://mlsanity.com/paper/2405.04097/) (2024)
- [Good practices for evaluation of synthesized speech](https://mlsanity.com/paper/2503.03250/) (2025)
- [Can Emotion Fool Anti-spoofing?](https://mlsanity.com/paper/2505.23962/) (2025)
- [Human perception of audio deepfakes: the role of language and speaking style](https://mlsanity.com/paper/2512.09221/) (2025)
- [Every Breath You Don't Take: Deepfake Speech Detection Using Breath](https://mlsanity.com/paper/2404.15143/) (2024)
- [Can You Tell It's AI? Human Perception of Synthetic Voices in Vishing Scenarios](https://mlsanity.com/paper/2602.20061/) (2026)

## Abstract (arXiv)

Automatic deepfake detection has received considerable research attention, yet the socio-technical environment in which humans actually encounter synthetic speech remains poorly understood. We investigate voice deepfake detection as a perceptual and contextual process, presenting a localization task in which 47 participants marked suspected synthetic segments across authentic, fully synthetic, and partially synthetic utterances under three manipulated trust cues: instructional framing, affective priming, and provenance labeling. Participants provided quality ratings on mechanicalness, expressiveness, intelligibility, clarity, calmness, and confidence of evaluation. Utterance class was the primary determinant of detection accuracy and perceptual quality; trust cues produced no main effects but motivated detection behavior. Fully synthetic speech was detected at below-chance levels. Quality ratings tracked utterance type, indicating implicit discrimination where overt detection failed.