MODEL ID minimax:speech@2.8

live

MiniMax Speech 2.8

by MiniMax January 29, 2026

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.

API Options

Platform-level options for task execution and delivery.

taskType string required value: audioInference: Identifier for the type of task being performed

taskUUID string required UUID v4: UUID v4 identifier for tracking tasks and matching async responses. Must be unique per task.

outputType string default: URL: Audio output type.

Allowed values 3 values

outputFormat string default: MP3: File format for the generated audio.

Allowed values 3 values

webhookURL string URI

Specifies a webhook URL where JSON responses will be sent via HTTP POST when generation tasks complete. For batch requests with multiple results, each completed item triggers a separate webhook call as it becomes available.

Learn more 1 resource

Webhooks
PLATFORM

deliveryMethod string default: sync

Determines how the API delivers task results.

Allowed values 2 values

: Returns complete results directly in the API response.
: Returns an immediate acknowledgment with the task UUID. Poll for results using getResponse.

Learn more 1 resource

Task Responses
PLATFORM

uploadEndpoint string URI

Specifies a URL where the generated content will be automatically uploaded using the HTTP PUT method. The raw binary data of the media file is sent directly as the request body. For secure uploads to cloud storage, use presigned URLs that include temporary authentication credentials.

Common use cases:

Cloud storage: Upload directly to S3 buckets, Google Cloud Storage, or Azure Blob Storage using presigned URLs.
CDN integration: Upload to content delivery networks for immediate distribution.

// S3 presigned URL for secure upload
https://your-bucket.s3.amazonaws.com/generated/content.mp4?X-Amz-Signature=abc123&X-Amz-Expires=3600

// Google Cloud Storage presigned URL
https://storage.googleapis.com/your-bucket/content.jpg?X-Goog-Signature=xyz789

// Custom storage endpoint
https://storage.example.com/uploads/generated-image.jpg

The content data will be sent as the request body to the specified URL when generation is complete.

ttl integer min: 60: Time-to-live (TTL) in seconds for generated content. Only applies when outputType is 'URL'.

includeCost boolean default: false: Include task cost in the response.

numberResults integer min: 1 max: 20 default: 1: Number of results to generate. Each result uses a different seed, producing variations of the same parameters.

Generation Parameters

Core parameters for controlling the generated content.

model string required value: minimax:speech@2.8

Identifier of the model to use for generation.

Learn more 3 resources

audioSettings object

Advanced settings for audio generation.

Properties 3 properties

audioSettings » bitrate bitrate integer min: 1 default: 128: Target audio bitrate in kbps. Constraints depend on the output format.

Allowed values 4 values

audioSettings » sampleRate sampleRate integer min: 1000 default: 32000: Audio sample rate in Hz.

Allowed values 6 values

audioSettings » channels channels integer default: 2: Number of audio channels in the generated output (1 for Mono, 2 for Stereo).

Allowed values 2 values

speech object required

Settings for speech generation.

Properties 2 properties

speech » text text string required max: 50000: Text to convert to speech.

speech » voice voice string required: Voice identifier from the Minimax voice library.

Allowed values 332 values

Settings

Technical parameters to fine-tune the inference process. These must be nested inside the settings object.

settings » languageBoost languageBoost string default: auto: Language detection and override for Speech synthesis.

Allowed values 41 values

settings » turbo turbo boolean default: false: Enable faster, lower-cost generation with potentially reduced quality.

taskType string required value: audioInference: Type of the task.

taskUUID string required UUID v4: UUID of the task.

audioUUID string required UUID v4: UUID of the output audio.

audioURL string URI: URL of the output audio.

audioBase64Data string: Base64-encoded audio data.

audioDataURI string URI: Data URI of the output audio.

seed integer: The seed used for generation. If none was provided, shows the randomly generated seed.

cost float: Task cost in USD. Present when includeCost is set to true in the request.

Text to Audio

Urgent Alpine Funicular Bulletin

0:00

Request

{
  "taskType": "audioInference",
  "taskUUID": "8ec44723-8939-44ff-986c-4f91d5d06159",
  "model": "minimax:speech@2.8",
  "speech": {
    "voice": "French_Female_News Anchor",
    "text": "Attention, ici le poste de régulation du funiculaire de Saint-Clair. En raison d’un glissement de terrain signalé au-dessus de la voie nord, le service est interrompu jusqu’à nouvel ordre. Les passagers déjà arrivés à la station supérieure sont priés de rester à l’abri dans le hall principal et de suivre les consignes du personnel. Une navette routière partira de l’esplanade est dans quinze minutes pour assurer la descente vers la vallée. Nous vous demandons de ne pas emprunter les sentiers balisés autour de la crête, car des chutes de pierres restent possibles. Les familles avec enfants, les personnes âgées et toute personne ayant besoin d’assistance seront prises en charge en priorité. Un nouveau point d’information sera diffusé à dix-sept heures quarante-cinq. Merci de votre calme et de votre coopération."
  },
  "settings": {
    "languageBoost": "fr",
    "turbo": false
  }
}

Response

{
  "taskType": "audioInference",
  "taskUUID": "8ec44723-8939-44ff-986c-4f91d5d06159",
  "audioUUID": "ba903c2d-3679-4505-bb58-001c569d0d85",
  "audioURL": "https://am.runware.ai/audio/os/a04d20/ws/5/ai/ba903c2d-3679-4505-bb58-001c569d0d85.mp3",
  "cost": 0.0819
}

Text to Audio

Nocturnal Museum Heist Monologue

0:00

Request

{
  "taskType": "audioInference",
  "taskUUID": "40ca27df-a334-4593-acc4-8e8fea104837",
  "model": "minimax:speech@2.8",
  "speech": {
    "voice": "Spanish_CaptivatingStoryteller",
    "text": "A las dos y trece de la madrugada, el museo parecía dormido, pero yo sabía que estaba escuchando. Cada vitrina respiraba silencio; cada pasillo guardaba un eco distinto, como si el edificio recordara los pasos de todos los que alguna vez soñaron con tocar lo intocable. No vine por oro, ni por fama, ni siquiera por venganza. Vine por una caja pequeña, forrada en nogal oscuro, escondida detrás de un reloj solar del siglo diecisiete. Dentro, según decían, había una carta que no debía existir. La escribió una reina en su última noche de libertad. Y esa carta, si era auténtica, podía cambiar la historia de un país entero. Cuando la alarma láser rozó la punta de mi guante, contuve el aliento. Un movimiento más, y todo habría terminado. Entonces oí una voz en la penumbra: 'Si has llegado hasta aquí, más te vale saber leer secretos'. No era un guardia. No era una grabación. Era alguien que me había estado esperando."
  },
  "settings": {
    "languageBoost": "es",
    "turbo": false
  }
}

Response

{
  "taskType": "audioInference",
  "taskUUID": "40ca27df-a334-4593-acc4-8e8fea104837",
  "audioUUID": "7777ed63-000b-4c45-a815-3e50c6b74ad5",
  "audioURL": "https://am.runware.ai/audio/os/a11d13/ws/5/ai/7777ed63-000b-4c45-a815-3e50c6b74ad5.mp3",
  "cost": 0.0921
}

Text to Audio

Midnight Museum Evacuation Announcement

0:00

Request

{
  "taskType": "audioInference",
  "taskUUID": "4d624adb-6017-4e06-9dbc-c756f05f8cbc",
  "model": "minimax:speech@2.8",
  "speech": {
    "voice": "English_expressive_narrator",
    "text": "Attention all remaining guests. This is the final announcement for Hall C. The building is closing early due to an unexpected power fluctuation in the east wing. Please remain calm and follow the blue floor lights to the nearest exit. Staff members in silver badges will guide you through the sculpture gallery and down the west staircase. If you are separated from your group, stop at the marble fountain in the central atrium and wait for assistance. Do not use the elevator. I repeat, do not use the elevator. The exhibits are secure, the situation is under control, and your safety is our highest priority. Thank you for your patience, and please proceed carefully."
  },
  "settings": {
    "languageBoost": "en",
    "turbo": false
  }
}

Response

{
  "taskType": "audioInference",
  "taskUUID": "4d624adb-6017-4e06-9dbc-c756f05f8cbc",
  "audioUUID": "7115ca25-5dee-4fc2-8029-4bce7aa931d4",
  "audioURL": "https://am.runware.ai/audio/os/a08d21/ws/5/ai/7115ca25-5dee-4fc2-8029-4bce7aa931d4.mp3",
  "cost": 0.0669
}

Text to Audio

Subglacial Rescue Radio Bulletin

0:00

Request

{
  "taskType": "audioInference",
  "taskUUID": "cf5dd067-f19a-4b68-bb78-7c8f2c4c506f",
  "model": "minimax:speech@2.8",
  "speech": {
    "voice": "English_Trustworth_Man",
    "text": "Attention all survey teams, this is Base Relay Seven broadcasting on emergency channel two. At 04:13 station time, a pressure ridge opened beneath the eastern ice shelf and cut power to the remote drilling camp. If you are hearing this message, do not attempt the old supply tunnel. Repeat: do not enter the old supply tunnel. Thermal scans show shifting fractures, falling particulate ice, and rapidly dropping oxygen levels in the lower passage. Rescue crews are moving in from the west with portable heaters, medical packs, and guide beacons. Team Aurora, hold your current position near the blue anchor line and conserve battery power. Team Calder, mark every junction with reflective tape and answer each beacon ping with three clicks. If visibility fails, stay low, stay linked, and count your breathing. We have partial contact with two missing engineers. Their last transmission reported a humming sound behind the wall of the melt chamber, followed by warm air and running water where no water should be. Listen carefully: if you encounter fog, metallic dust, or a repeating whistle that does not match your equipment, stop immediately and report your coordinates. You are not alone down there. Keep your radios open. Rescue is fourteen minutes out."
  },
  "settings": {
    "languageBoost": "en",
    "turbo": false
  }
}

Response

{
  "taskType": "audioInference",
  "taskUUID": "cf5dd067-f19a-4b68-bb78-7c8f2c4c506f",
  "audioUUID": "52117d8f-254f-48de-bea4-a77e12511ecf",
  "audioURL": "https://am.runware.ai/audio/os/a11d13/ws/5/ai/52117d8f-254f-48de-bea4-a77e12511ecf.mp3",
  "cost": 0.1258
}