Voice and speech

MiniMax Speech 2.8 (HD and Turbo)

Name: MiniMax Speech 2.8 (HD and Turbo)
Author: MiniMax (Hailuo AI)

MiniMax's current expressive TTS line, announced January 2026, in an HD variant for quality and a Turbo variant for speed.

A speech model by MiniMax (Hailuo AI) · read 2026-08-14 · sourcing low

Resolution: up to 44.1kHz
Weights: Not released
API: Yes
Commercial use: Permitted subject to conditions

It supports roughly 40 languages via a language_boost parameter with auto-detection, named emotion presets, pitch/intensity/timbre sliders, and sound tags like laughs and sighs. Rapid voice cloning works from around 10 seconds of audio, and Voice Design creates voices from text prompts.

Commercial use — Permitted subject to conditions

Do not assume cloned-voice commercial rights without reading MiniMax's own terms.

Requires consent from the person depicted. Consent and verification requirements for Rapid Voice Cloning could not be verified - the MiniMax platform documentation and pricing pages returned 404 or empty content on fetch

The conditions above were extracted from the licence notes recorded during research, and each one shows the sentence it came from so you can check the reading. Read the licence before you rely on any of this. Nothing on this page is legal advice, and this site does not tell you a model is cleared for commercial use.

Specification

Vendor figures, read 2026-08-14Read from MiniMax (Hailuo AI)'s own documentation on 2026-08-14. Vendors change these between model versions without announcing it, so check theirs before you commit a shot to a number. Sourcing for this entry is thin. Treat it as a starting point and verify before relying on any figure. Check the vendor's current documentation

What it takes in: text, audio
What it makes: audio
Maximum resolution: up to 44.1kHz
Controls: emotion-control, style-prompting, voice-cloning, speed-control, streaming-low-latency
Weights: Not released
API: Yes
Pricing: Reported official pay-as-you-go: $100 per 1M characters for HD variants and $60 per 1M for Turbo; Rapid Voice Cloning $1.50 per voice; Voice Design $3.00 per created voice. Recorded from a third-party compilation citing MiniMax's pricing page - the vendor pricing page could not be fetched directly.
Released or updated: 2026-01-23 (Speech 2.8 announcement, per third-party report)
Languages: 40, not individually listed by the vendor

Strengths

Sound tags (laughs, sighs, coughs) for performance texture
~40 languages with automatic detection
10-second voice cloning and text-prompt voice design
Widely resold through third-party inference hosts, so easy to trial
Turbo at a reported $60/1M characters undercuts ElevenLabs' quality tier

Weaknesses

Vendor pricing and API documentation pages were unreachable at time of check - all pricing is second-hand
Max characters per request not documented anywhere reachable
Speech 2.6 was reportedly moved to 'Legacy Audio Models' around mid-2026, so version churn is fast
Consent policy for voice cloning unverified
Chinese vendor may raise procurement or data-residency questions for some productions

Best for

expressive character VO
multilingual short-form content
cost-sensitive expressive TTS

Where the sources disagree

Recorded rather than resolved. Picking the tidier number would hide the fact that the vendor's own pages do not agree.

Language count is reported as '40+' in MiniMax's own Speech 2.6 announcement and '~40 via language_boost' in third-party specs; recorded as 40.
Latency: MiniMax's Speech 2.6 announcement claims end-to-end under 250ms; no figure published for 2.8, so latency is not recorded for this entry.
Pricing is identical ($100 HD / $60 Turbo) for 2.6 and 2.8 in the third-party compilation, which is plausible but unconfirmed on a vendor page.

What to understand before using it

Explainers naming the mechanisms this model's own documentation mentions.

Character and set consistency: the weakest link names “character”
Camera control: a move without a camera names “control”
Conditioning and control: everything you hand the model besides noise names “control”

MCP servers that front MiniMax (Hailuo AI)

Servers in the media MCP directory whose own description or homepage names this vendor. Absence of a server here is not evidence that none exists: most publish no tool list at all.

Aetherwave Music
Music, image, video and audio generation across top AI providers - one key, one credit pool.
BlitzReels Video Editor Video generation
Create, edit, reframe, caption, organize, generate, and export short-form videos with BlitzReels.
Clipia Image generation
Generate AI images, video, speech, music and presentations from Claude, ChatGPT and Cursor.
Corent Image generation
Generate images and videos via the Corent API, auto model routing with provider fallback.
Creative Claw Image generation
On-brand creative studio for AI agents: images, video, audio, and 3D.
Distribea Image generation
Generate images, video, music and voice from your CLI or AI agent. On-brand AI media toolkit.

See it from the vendor

Shotcraft hosts no sample output. Judging a model means watching its own showreel with your own eyes, and a clip re-hosted here would be a clip we chose.

Nearby in the directory

Azure AI Speech text-to-speech (Neural, Neural HD, MAI voices)

Microsoft's enterprise speech synthesis service, with 200+ prebuilt neural voices across 80+ locales plus tiers for HD voices, multilingual voices, mu

Voice and speech

Sonic 3.5

Cartesia's current TTS model, documented at sub-90ms first-byte latency across 42 languages. It is tuned for conversational delivery and for reading a

Voice and speech

XTTS-v2

An open-weights multilingual TTS and voice cloning model covering 17 languages at 24kHz, still widely used for local voice cloning. Coqui Inc. shut do

Voice and speech

Sources

Read on 2026-08-14. A source that has changed since then will not be reflected here until the next pass.