MiniMax: MiniMax Speech 2.8

MiniMax Speech 2.8 is an advanced text-to-speech model that turns text into natural, expressive audio in multiple languages. It delivers broadcast-ready speech with rich prosody, emotional control, and a diverse voice library. The model supports up to large input lengths and can be used for voiceovers, narration, accessibility tools, and interactive voice applications.

Compare models

minimax-speech-2-8

AvailableMiniMaxText to speechSync

Input / Output: $100.00 / $0.00
Modalities: Audio

About MiniMax Speech 2.8

MiniMax Speech 2.8 is MiniMax's text-to-speech model family, released in January 2026, which turns text into expressive audio in many languages. It brings natural sound tags such as laughs and sighs, emotional control, and a large voice library. This is the general Speech 2.8 model; the HD and Turbo variants split the family by quality and speed.

Where it works well

Supports interjection tags written in the text, such as (laughs), (sighs) and (breath), per MiniMax.
Offers rich prosody and emotional control for narration and dialogue.
Covers many languages with a diverse library of system and cloned voices.
Speed, pitch and volume are adjustable, with mp3, pcm, flac and wav output.

When to choose another model

Speech only: it reads text aloud and does not transcribe or translate audio.
It is the all-round Speech 2.8 model; the HD variant is the choice for maximum fidelity and Turbo for lowest delay.

Getting started

Create API key
Create a key in Console, then use it with every model on the platform.

Send your first request

Copy the example for your language and run it against the endpoint.

Text to speechTokenLab endpoint

POST/v1/audio/speech

curl -X POST "https://api.tokenlab.sh/v1/audio/speech" \
  -H "Authorization: Bearer sk-xxx" \
  -H "Content-Type: application/json" \
  -d '{
  "operation": "tts",
  "model": "minimax-speech-2-8",
  "input": "Hello from TokenLab.",
  "voice": "alloy"
}'

Pricing

The Verified price applies to Verified, which costs less on most models. The Official price is the model maker's published price and applies to the more reliable Official route. Auto bills the route that completes the request.

Text to speech

per 1M characters

Official price: Input $100.00 / Output $0.00
Verified price: —
Discount: —

	Official priceper 1M characters	Verified priceper 1M characters	Discount
Text to speech	Input $100.00 / Output $0.00	—	—

Usage & activity

Success rate is the share of requests that completed. Latency is how long a full response takes; P95 means 95% of requests finished within that time.

Usage & availability

Last 24 hours

Requests
Success rate
P95 latency
Total tokens

Model performance

Last updated Oct 20, 2026, 6:44 PM UTC

Metrics appear once privacy and data volume thresholds are met.

Data is based on aggregate user requests, excluding status checks.

Open in Console

Open MiniMax Speech 2.8 in Console with a prompt ready to edit or send.

Help me try minimax-speech-2-8 with a short audio request at /v1/audio/speech. Show the result and cost.

Use cases

Voiceovers and narration
Turn scripts into audio for explainer videos, courses or audiobooks in the voice you choose.
Accessibility features
Read articles, notices and interface text aloud for people who prefer listening or have trouble reading on screen.
Interactive voice apps
Give game characters or assistants expressive lines with laughs and pauses written in.

Prompt examples

Welcome back. (breath) Today we are going to build something small, and it will take ten minutes.

(chuckle) I told you the map was upside down. Anyway, turn left at the old mill.

Your appointment is confirmed for Tuesday at three. (sighs) Please arrive ten minutes early.

FAQ

What is MiniMax Speech 2.8?

It is MiniMax's text-to-speech family that adds natural sound tags, lifelike voices and clean audio. This is the general model, with HD and Turbo variants offered separately.

Does it support laughter and breathing sounds?

Yes. MiniMax documents interjection tags, including laughs, chuckles, coughs, sighs, gasps and breaths, which you place in the text exactly where the sound should occur.

How is it different from speech-2.8-hd?

The HD variant is described for polished narration and character dialogue, where listening quality matters most. The general model covers the family broadly: many languages, emotion control and a diverse voice library.

Can I clone a voice?

MiniMax lists both system and cloned voices for its speech models. Choose a voice by its ID in the request; the voice library covers many languages and styles.

Which output formats are available?

MiniMax's speech API documents mp3, pcm, flac and wav output, with streaming over HTTP and WebSocket, which suits both long scripts and live playback in an app.

How much does MiniMax Speech 2.8 cost?

On TokenLab, MiniMax Speech 2.8 costs Input $100.00 / Output $0.00 per 1M characters. The pricing table above shows the full breakdown. Rates depend on the billing unit, specification, and usage. Compare matching conditions in the model's detailed pricing; a single rate does not determine the total cost.

Which endpoint should MiniMax Speech 2.8 use?

Use https://api.tokenlab.sh/v1/audio/speech for MiniMax Speech 2.8. The request example below shows the matching code shape.

Which operations does MiniMax Speech 2.8 support?

MiniMax Speech 2.8 supports Text to speech. Select an operation above to see its endpoint and request example.