OverblastMiniMax · speech model
Speech 2.8 HD
Speech 2.8 HD is a speech model on the Overblast AI Gateway. Call it as minimax/speech-2.8-hd with one Overblast API key, paid from your AI credits.
Prices are in AI credits, in US dollars. A picture, clip or track shows an estimate and is charged its actual cost once it is made. As of 2 October 2026, from the live catalogue.
Price
| Price | $0.20 / 1,000 characters |
|---|---|
| Least a call costs | $0.01 |
What it does
| Does | Speaks text aloud |
|---|---|
| Voices | 17: Wise_Woman, Friendly_Person, Inspirational_girl, Deep_Voice_Man, Calm_Woman, Casual_Guy, Lively_Girl, Patient_Man, Young_Knight, Determined_Man, Lovely_Girl, Decent_Boy, Imposing_Manner, Elegant_Man, Abbess, Sweet_Girl_2, Exuberant_Girl |
| Default voice | Wise_Woman |
| Audio formats | mp3 |
| Language | MUL |
| Longest input | 4,096 characters |
Where it runs
| Regions | Runs globally; it cannot be pinned to the EU or the US yet |
|---|---|
| Zero data retention | Not applicable: speech calls cannot carry the requirement |
| Endpoint | POST /v1/audio/speech |
Call it
curl https://ai.overblast.app/v1/audio/speech \
-H "Authorization: Bearer $OVERBLAST_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "minimax/speech-2.8-hd",
"input": "Hello from Overblast.",
"voice": "Wise_Woman",
"response_format": "mp3"
}' \
--output speech.mp3All speech models and their prices · How the AI Gateway works