# DesignArena Text to speech

> Source: Benchize (https://benchize.com) — https://benchize.com/benchmarks/designarena%3Atts. When citing or reusing this content, credit "Benchize" and link to https://benchize.com/benchmarks/designarena%3Atts.

Human pairwise preference evaluation of text to speech. Voters compare model outputs for the same request; published Elo ratings and separate model configurations are retained for this specific arena and input modality.

Canonical URL: https://benchize.com/benchmarks/designarena%3Atts
Last updated: Wed, 21 Oct 2026 18:52:20 GMT

Reproduction and quotation must name Benchize, link to the canonical HTML URL, and retain individual source citations.

Human pairwise preference evaluation of text to speech. Voters compare model outputs for the same request; published Elo ratings and separate model configurations are retained for this specific arena and input modality.

Multimodal · Models tested: 2 · Score publishers: [designarena](https://www.designarena.ai/leaderboard/tts)

## Model results

| # | Model | Score |  |
| --- | --- | --- | --- |
| 1 | [Gemini 3.8 Flash TTS](https://benchize.com/models/gemini-3-8-flash-tts.md) | 1,452elo |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `elo` \| 1452 \| elo \| tts \| Higher is better \| ### Configuration arenaType models category tts inputModality Not reported sourceModelVersion google_gemini_3_8_flash_tts variationName public ### Reported statistics averageGenerationTimeMs 20612 battles 2592 losses 579 standardError Not reported winRate 77.7 wins 2013 ### Source [Original source](https://www.designarena.ai/api/leaderboard#board=tts) Source designarena Original model label Google Gemini 3.8 Flash TTS Source version public Observed 2026-10-21T16:38:08.030Z Evidence SHA-256 `6d5bb46f625a899a9b8f42c39b8b14d82c8230c1e180f20b67c724e4c20fa4dd` Source locator `/data/0` |  |  |  |
| 2 | [Voxtral Mini TTS](https://benchize.com/models/voxtral-mini-tts-2603.md) | 993elo |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `elo` \| 993 \| elo \| tts \| Higher is better \| ### Configuration arenaType models category tts inputModality Not reported sourceModelVersion voxtral-mini-tts-2603 variationName public ### Reported statistics averageGenerationTimeMs 7235 battles 3142 losses 2240 standardError Not reported winRate 28.7 wins 902 ### Source [Original source](https://www.designarena.ai/api/leaderboard#board=tts) Source designarena Original model label Voxtral Mini TTS (2603) Source version public Observed 2026-10-21T16:38:08.030Z Evidence SHA-256 `6d5bb46f625a899a9b8f42c39b8b14d82c8230c1e180f20b67c724e4c20fa4dd` Source locator `/data/25` |  |  |  |

---
Cite as: DesignArena Text to speech — Benchize, https://benchize.com/benchmarks/designarena%3Atts (retrieved 2026-10-21)