Skip to content
FluxCDNFluxCDN
esc

MiniMax: Speech 2.8 HD

Text to speech

MiniMax Speech 2.8 HD is a text-to-speech model from MiniMax. It is suited for applications that generate spoken audio from text and accepts arbitrary MiniMax voice IDs.

Price
36,039 Toman per 1,000 characters
Endpoint
POST /v1/audio/speech
Published
Jul 16, 2026
Offers
45 voices
Offers
MP3
Maker
MiniMax

Free plan prices, in Toman at today's rate. Plans and prices

Price on each plan

Per character of the text read, in Toman at today's rate; a text of 1,000 characters is about a minute and a half of speech.

PlanPrice
Free36,039 Toman per 1,000 characters
Pro34,562 Toman per 1,000 characters
Max33,085 Toman per 1,000 characters

Voices

  • English_expressive_narratorExpressive narrator
  • English_CalmWomanCalm
  • English_Trustworth_ManTrustworthy
  • English_FriendlyPersonFriendly

45 voices in all: any of them may be asked for by name in the API.

Hear all voices

Call it

curl https://ai.fluxcdn.cloud/api/v1/audio/speech -H "Authorization: Bearer $FLUXCDN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "minimax/speech-2.8-hd", "input": "سلام! به FluxCDN خوش آمدید.", "voice": "English_expressive_narrator", "response_format": "mp3"}' \
  --output speech.mp3