---
type: source
id: lavan-2025-voice-clones-hyperrealism
title: Voice clones sound realistic but not (yet) hyperrealistic
authors:
- Nadine Lavan
- Mairi Irvine
- Victor Rosi
- Carolyn McGettigan
year: 2025
venue: PLOS ONE
volume: 20(9)
pages: e0332692
doi: 10.1371/journal.pone.0332692
peer_review: peer-reviewed-journal
evidence_tier: B
design: 'Three online experiments: generic AI voices vs human (Exp1a), voice clones vs the same humans (Exp1b), both together (Exp2); realness, human-vs-AI classification, trustworthiness and dominance ratings'
modality:
- audio
stimuli: 40 VCTK speakers; 40 ElevenLabs Instant Voice Cloning clones (≈4 min of training audio each); 40 ElevenLabs Voice Design generic voices; one Rainbow Passage sentence each, artefact-screened
n_participants: 150
n_note: n=50 per experiment; UK native English speakers via Prolific; not preregistered
topics:
- human-detection
- voice
- voice-clone
- hyperrealism
- trust
summary: Listeners could not tell voice clones from the real speakers (d' 0.10-0.13) and rated them as real as human voices, but clones were not judged more human than humans (no hyperrealism); AI voices were rated more dominant.
results:
- metric: 'Exp1b proportion labelled human: clones vs real'
  value: 0.58 vs 0.62
  note: OR = 0.84, p = .239; d' = 0.10, n.s.
  source_location: null
- metric: 'Exp2 proportion labelled human: clones vs real'
  value: 0.70 vs 0.72
  note: d' clones 0.13; generic AI voices 0.39 vs 0.72, d' 2.93
  source_location: null
- metric: 'Exp2 realness ratings: clones vs real'
  value: 63.0 vs 63.8
  note: p = .682
  source_location: null
- metric: 'Dominance: clones vs real'
  value: 49.4 vs 44.0
  note: p = .015; generic AI 58.1 vs 40.4
  source_location: null
- metric: 'Trustworthiness: clones vs real'
  value: 58.9 vs 57.1
  note: n.s.; generic AI voices 57.9 vs 53.8, p = .002
  source_location: null
moderators:
- Generic (non-cloned) AI voices were easy to spot; clones were not
- Criterion biased toward 'human' for clones (c = -0.65 in Exp2)
limitations:
- Unfamiliar voices only
- Artefact-screened, idealised stimuli
- One sentence per voice
- Not preregistered
- Context effects across experiments
verification:
  status: verified-fulltext
  confidence: high
  checked: 2026-10-12
  via: journals.plos.org full text
url: null
pmid: null
pmcid: null
arxiv_id: null
openalex_id: null
oa_url: null
cited_in:
- MR-0001
- MR-0002
- MR-0003
updated: '2026-10-12'
---

# Voice clones sound realistic but not (yet) hyperrealistic

> **Source record** `lavan-2025-voice-clones-hyperrealism` · 2025 · PLOS ONE  
> **Evidence tier:** B (peer-reviewed-journal) · **Verification:** verified-fulltext  
> **Bottom line:** Listeners could not tell voice clones from the real speakers (d' 0.10-0.13) and rated them as real as human voices, but clones were not judged more human than humans (no hyperrealism); AI voices were rated more dominant.

## Citation {#citation}

Nadine Lavan, Mairi Irvine, Victor Rosi, Carolyn McGettigan (2025). Voice clones sound realistic but not (yet) hyperrealistic. *PLOS ONE*, 20(9), e0332692. https://doi.org/10.1371/journal.pone.0332692

Verification entry (identifiers, resolvers, claims with source locations): [/data/citations.json#lavan-2025-voice-clones-hyperrealism](/data/citations.json#lavan-2025-voice-clones-hyperrealism)

## Study design {#design}

| Field | Value |
|---|---|
| Design | Three online experiments: generic AI voices vs human (Exp1a), voice clones vs the same humans (Exp1b), both together (Exp2); realness, human-vs-AI classification, trustworthiness and dominance ratings |
| Modality | audio |
| Stimuli | 40 VCTK speakers; 40 ElevenLabs Instant Voice Cloning clones (≈4 min of training audio each); 40 ElevenLabs Voice Design generic voices; one Rainbow Passage sentence each, artefact-screened |
| Participants (N) | 150 |
| N detail | n=50 per experiment; UK native English speakers via Prolific; not preregistered |
| Topics | human-detection, voice, voice-clone, hyperrealism, trust |

## Results {#results}

| Metric | Value | Note |
|---|---|---|
| Exp1b proportion labelled human: clones vs real | 0.58 vs 0.62 | OR = 0.84, p = .239; d' = 0.10, n.s. |
| Exp2 proportion labelled human: clones vs real | 0.70 vs 0.72 | d' clones 0.13; generic AI voices 0.39 vs 0.72, d' 2.93 |
| Exp2 realness ratings: clones vs real | 63.0 vs 63.8 | p = .682 |
| Dominance: clones vs real | 49.4 vs 44.0 | p = .015; generic AI 58.1 vs 40.4 |
| Trustworthiness: clones vs real | 58.9 vs 57.1 | n.s.; generic AI voices 57.9 vs 53.8, p = .002 |

## Moderators {#moderators}

- Generic (non-cloned) AI voices were easy to spot; clones were not
- Criterion biased toward 'human' for clones (c = -0.65 in Exp2)

## Limitations {#limitations}

- Unfamiliar voices only
- Artefact-screened, idealised stimuli
- One sentence per voice
- Not preregistered
- Context effects across experiments

## Cited in {#cited-in}

- [MR-0001](/metareports/human-detection-accuracy.md): Human Detection of Deepfakes: Accuracy, Bias, Confidence, and Training
- [MR-0002](/metareports/uncanny-valley-and-synthetic-faces.md): The Uncanny Valley and Synthetic Media: Does It Exist, What Causes It, and Have Deepfakes Crossed It?
- [MR-0003](/metareports/voice-cloning-human-detection.md): Voice Clones and Synthetic Speech: How Well Do People Detect Them?

## Verification {#verification}

| Field | Value |
|---|---|
| Status | `verified-fulltext` |
| Compiler confidence | high |
| Checked on | 2026-10-12 |
| Via | journals.plos.org full text |