MiniMax: MiniMax Speech 2.8 Turbo
speech-2.8-turbo- Input / Output
- $60.00 / $0.00
- Modalities
- Audio
About MiniMax Speech 2.8 Turbo
MiniMax Speech 2.8 Turbo is the speed-focused text-to-speech model in the Speech 2.8 line. It voices a script in a selected voice with lower delay, which makes it a fit for responsive spoken interfaces and quick narration. It supports the same interjection sound tags as the HD model; choose HD when finished listening quality matters more than turnaround.
Where it works well
- Built around speed, so replies in a voice interface start sooner.
- Supports interjection tags like (laughs) and (breath) in the text.
- Works with the same system and cloned voices as other Speech 2.8 models.
- Streaming output over HTTP and WebSocket is documented for MiniMax's speech API.
When to choose another model
- For published audiobook or ad narration, the HD model is described for higher polish.
- It converts text to speech only; it cannot recognize or edit existing recordings.
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 endpointPOST/v1/audio/speechcurl -X POST "https://api.tokenlab.sh/v1/audio/speech" \ -H "Authorization: Bearer sk-xxx" \ -H "Content-Type: application/json" \ -d '{ "operation": "tts", "voice": "male-qn-qingse", "model": "speech-2.8-turbo", "input": "Hello from TokenLab." }'
Parameters
Text to speech
| Parameter | Values and limits | Default | Notes |
|---|---|---|---|
input | Not provided | Not provided | Not provided |
model | Not provided | Not provided | Not provided |
response_format | mp3, wav, pcm, flac | Not provided | Not provided |
speed | ≥ 0.5, ≤ 2 | Not provided | The gateway accepts numbers from 0.25 to 4.0. The selected supplier can impose a narrower range. A uniform omitted-field default is not verified. |
stream_format | audio | audio | Not provided |
voice | Not provided | Not provided | This field is accepted. The current public contract does not specify a complete value range or list. No additional values are promised. The omitted-field default is not verified. |
voice_id | Not provided | Not provided | This field is accepted. The current public contract does not specify a complete value range or list. No additional values are promised. The omitted-field default is not verified. |
Selected voice, speed and audio format; no unsupported synthesis controls are promised.
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.
Pricing
per 1M characters- Verified price
- —
- Official price
- Input$60.00/1M charactersOutput$0.00/1M characters
| Official priceper 1M characters | Verified priceper 1M characters | Discount | |
|---|---|---|---|
| Input | $60.00 | — | — |
| 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
Data is based on aggregate user requests, excluding status checks.
Open in Console
Open MiniMax Speech 2.8 Turbo in Console with a prompt ready to edit or send.
Help me try speech-2.8-turbo with a short audio request at /v1/audio/speech. Show the result and cost.
Use cases
Voice assistants
Speak answers aloud in a chat or phone assistant without a noticeable pause before audio starts.
Quick narration drafts
Hear a script read in a chosen voice while it is still being revised.
In-app read-aloud
Add a listen button to articles, notifications or learning material so readers can hear the content instead of reading it.
Prompt examples
Your order has shipped. (breath) It should reach you by Thursday, and I will text you when it is close.
Good morning! Here are today's three headlines, starting with the weather.
(chuckle) Sure, let me look that up for you. One moment.
FAQ
What is speech-2.8-turbo for?
It is MiniMax's faster Speech 2.8 variant, aimed at responsive spoken interfaces and narration where turnaround matters. It converts a script to audio in a voice you select.
Does Turbo sound worse than HD?
MiniMax positions HD for ultra-realistic output and Turbo for speed. The gap depends on voice and script, so listen to both with your own text before choosing.
Can I use sound tags with Turbo?
Yes. MiniMax lists interjection tags as supported on both speech-2.8-hd and speech-2.8-turbo, including laughs, sighs, breaths and coughs, written in parentheses within the script text.
Is it good for live conversation?
It suits low-delay use better than HD does, and MiniMax documents streaming over HTTP and WebSocket. Measure end-to-end delay in your own stack, since network and text length matter.
Which formats can it output?
MiniMax's speech API documents mp3, pcm, flac and wav output. Use mp3 for compact web playback and wav or flac when the audio will be edited further.
How much does MiniMax Speech 2.8 Turbo cost?
On TokenLab, MiniMax Speech 2.8 Turbo costs Input $60.00 / Output $0.00 per 1M characters. The pricing table above shows the full breakdown.
Which endpoint should MiniMax Speech 2.8 Turbo use?
Use https://api.tokenlab.sh/v1/audio/speech for MiniMax Speech 2.8 Turbo. The request example below shows the matching code shape.
Which operations does MiniMax Speech 2.8 Turbo support?
MiniMax Speech 2.8 Turbo supports Text to speech. Select an operation above to see its endpoint and request example.
Compare MiniMax Speech 2.8 Turbo
Sources
Reviewed Oct 2, 2026
