Seedance 2.5 is here. Every model is included on every plan.
PricingLog inSign up
VoiceMiniMax

MiniMax Speech-02 HD

MiniMax Speech-02 HD is MiniMax's studio voice model, with a compact character voice set and broad language coverage. Reach for it when a script has to work in several languages with a consistent character.

Voices17 available
Make with MiniMax Speech-02 HD →View pricing

Added July 2026. Included on every plan.

Why this one

What makes MiniMax Speech-02 HD different

  1. Voices with characterThe set is cast by personality rather than by neutral tone, which suits brand and character work.
  2. Broad language coverageBuilt for multilingual delivery from the same voice identity.
  3. HD outputA high definition tier, which shows on headphones and in quiet rooms.
  4. A short, decisive voice listSeventeen voices, each distinct enough to choose between quickly.
  5. Per character pricingScript length is the cost, so estimates are simple.
  6. Descriptive voice namesThe names describe the delivery, which makes casting faster than auditioning blind.

Use cases

Best for

  • Brand voice
  • Character narration
  • Localisation
  • Advertising reads
  • Personality driven explainers
  • Multilingual campaigns

Technique

Prompt tips

  1. Cast by nameThe voice names describe delivery. Read the list before you audition and you will shortlist faster.
  2. Match the script to the characterA voice cast as lively fights a somber script. Pick the voice for the copy you actually have.
  3. Punctuate deliberatelyPacing follows punctuation here as on every voice model.
  4. Keep one voice across a campaignThe value of a character voice is recognition, which only accrues if you reuse it.
  5. Test the language you will ship inDelivery varies between languages. Audition in the target language, not in English.

Mechanism

How MiniMax Speech-02 HD works

You pick one of its voices and send the script, and it renders high definition speech billed per thousand characters. Its voices are cast rather than neutral, which means the model behaves less like a text to speech engine and more like a small roster of performers to choose between.

Getting started

How to use it on Dolly

  1. Pick MiniMax Speech-02 HDChoose it in the voice studio, or just ask the agent for it by name.
  2. Write the promptDescribe what you want, in as much detail as the brief deserves.
  3. GenerateThe exact credit cost sits on the button before you press it, and a failed run refunds automatically.

Pricing

What it costs

Every generation shows its exact credit cost on the button before you run it.

per 1,000 characters
20 credits

One run is about 1/63 of a Starter month.

Failed generations refund automatically. See plans

Capabilities

Specs

Voices
17

Questions

MiniMax Speech-02 HD FAQ

How many voices are there?
Seventeen, listed on this page with their names, and the names describe the delivery so you can shortlist before auditioning.
How many languages does it support?
Broad multilingual coverage from the same voice identities. Audition in your target language before committing.
How is it priced?
Per thousand characters of script, so the cost of a read follows its length rather than its duration.
Is it expressive?
Yes, more so than the neutral studio models, because the voices are cast by personality.
Can it do dialogue?
Render each speaker separately and assemble, or use ElevenLabs V3 for multi voice in one generation.

More like this

Other voice models

ElevenLabs Multilingual V2ElevenLabs Multilingual V2 is the classic ElevenLabs voice model, with the largest voice library on Dolly and consistently lifelike delivery.ElevenLabs V3ElevenLabs V3 is the expressive ElevenLabs model, capable of multi voice dialogue in a single generation.Gemini 2.5 Pro TTSGemini 2.5 Pro TTS is Google's studio quality speech model, with a curated voice set and multi speaker scene support.

Ready to make with MiniMax Speech-02 HD?

Every model is included on every plan. The price sits on the button before you run it.

Start making →