# HeyGen Voice Benchmark Scores & Performance

> HeyGen Voice has HeyGen voice studies results. Voice evidence is display-only and does not produce a text-model score.

Canonical page: https://benchlm.ai/models/heygen-voice

Last updated: October 14, 2026

## Model Details

| Property | Value |
|----------|-------|
| Creator | HeyGen |
| Source Type | Proprietary |
| Reasoning Type | Non-Reasoning |
| Context Window | N/A |
| Official model card | [HeyGen model card](https://dynamic.heygen.ai/www/research/introducing-heygen-voice/heygen-voice-scene-aware-voice-cloning-for-talking-videos.pdf) |
| Overall Score | Not scored (voice evidence only) |
| Overall Rank | Unranked |

## Family & Coverage

- Family: HeyGen Voice
- Variant: base
- Benchmarks covered: 0 of 629
- Coverage note: Speech and voice results appear below; no weighted text-model benchmark rows are stored.

## Speech and voice evidence

[HeyGen Voice evaluation](/voice-benchmarks/heygen-voice-evaluation)

HeyGen ran the digital-twin and platform studies. The 43-person ratings and 6,153-vote preference shares describe different studies and model modes. These observations do not enter weighted text-model scores. The research publication date does not establish first availability.

| Metric | Result |
| --- | --- |
| Digital-twin voice rating · voice-clone | 76.1/100 |
| Digital-twin voice rating · full-clone | 72.6/100 |
| Three-way clone preference · full-clone | 41.7% of 6,153 votes |

HeyGen-reported studies. Digital-twin ratings use 43 volunteers; platform preference votes are a separate study. Research publication date, not verified first availability.

- [Evaluation source](https://www.heygen.com/research/heygen-voice-scene-aware-voice-cloning)

HeyGen published its research report on October 14, 2026 and links a platform trial. The model combines a script, voice reference, target image, and scene descriptions to produce 44.1 kHz audio. Voice-clone and full-clone are modes of the same model; a separate avatar model renders the video. The report does not identify a versioned public API model ID. The report date is not a verified first-availability date.