MiniMax Speech 2.8 HD

Endpoint

POST https://api.roasos.com/v1/audio/minimax-speech-hd

Send your API key as a bearer token. The key needs the scope api:write. Every paid request also needs an Idempotency-Key header.

The envelope fields quote_only, quote_id, webhook_url and wait_timeout_seconds behave the same on each model endpoint. Refer to Jobs and Quotes and credits.

Good at

The multilingual workhorse: 239 presets, 24 languages, and it holds an accent in every one of them. HD is the studio-quality tier; use Turbo for drafts and long scripts. It is one of two models that accept a cloned voice_id from POST /v1/voices, so it is the choice for a brand voice that must stay the same across many ads. Pause syntax <#x#> and interjection tags such as (laughs) give a natural read. Set language when the text is not English. It accepts no image, video, or audio media; the clone sample belongs to /v1/voices.

Pricing

The unit is one characters. Pricing key: minimax-speech-hd.

The credits come from the roasOS pricing table. No other field changes the price. To get the exact price of your request, send it with quote_only: true first.

Modes

• text-to-speech

Provider endpoints: fal-ai/minimax/speech-2.8-hd.

Request fields

The schema is strict. A field that is not in this table gets 400 unsupported_option. The field names are the roasOS role names. One role has one name in every family.

Media

This endpoint accepts these media roles:

• voice_id — a saved voice clone

Every other role is refused: avatar_id, elements, product_id, product_images, references, reference_images, image_url, end_image_url, scene_id, scene_image, video_urls, video_url, audio_urls, audio_url. A refused role gets 400 unsupported_option.

Prompt

text limit 10,000 characters. roasOS adds no wrapper; the trimmed text is sent unchanged.

• Pause syntax: <#1.5#> inserts a 1.5 s pause (0.01–99.99).
• Interjection tags: (laughs), (sighs), (coughs), (clears throat), (gasps), (sniffs), (groans), (yawns).
• Set language to the language of the text; it boosts recognition. Mixed-language text: pick the dominant one.
• Use emotion for a whole-clip mood; use punctuation and tags for a local beat.
• Set english_normalization: true for prices, dates, and phone numbers in English.
• Keep speed between 0.9 and 1.15 for ad reads; large changes sound synthetic.

Not exposed

These catalog entries describe omitted inputs, pinned values and source constraints. They do not imply that the provider accepts every listed field.

• voice_setting.vol — pinned to 1.
• output_format — pinned to url.
• audio_setting — pinned to sample_rate, bitrate, format and channel.
• pronunciation_dict — not accepted by this public endpoint.
• normalization_setting — pinned to enabled, target_loudness, target_range and target_peak.
• voice_modify — pinned to pitch, intensity and timbre, each -100 to 100.
• language_boost — these values are not available: Nynorsk, auto.
• references — not accepted by this public endpoint.
• avatar_id — not accepted by this public endpoint.
• product_id — not accepted by this public endpoint.
• audio_urls — not accepted by this public endpoint.
• image_url — not accepted by this public endpoint.
• temperature — not accepted by this public endpoint.
• style_instruction — not accepted by this public endpoint.

Result

The result kind is audio. Always present: url. Present when the run makes it: timestamps.

{
  "data": {
    "id": "gen_…",
    "kind": "audio",
    "model": "minimax-speech-hd",
    "status": "completed",
    "result": {
      "timestamps": [
        {}
      ],
      "url": "https://cdn.roasos.com/workspaces/ws_01/result"
    }
  }
}

The files are immutable objects on https://cdn.roasos.com. A result URL is a valid media input for your next request.

Example

The three steps are always the same: quote, run, poll.

# 1. Quote. Nothing is spent.
curl -X POST https://api.roasos.com/v1/audio/minimax-speech-hd \
  -H "Authorization: Bearer roasos_sk_live_..." \
  -H "Idempotency-Key: 0f1c9e2a-minimax-speech-hd-0001" \
  -H "Content-Type: application/json" \
  -d '{"quote_only":true,"text":"Two drops, morning and night. That is the whole routine."}'
# 200 { "data": { "kind": "quote", "quote_id": "q_7Hk...", "credits": 2, "expires_at": "...",
#        "confirmation": { "model": "minimax-speech-hd", "example": "1000_characters" } } }

# 2. Run. Same body, same Idempotency-Key, quote_id in place of quote_only.
curl -X POST https://api.roasos.com/v1/audio/minimax-speech-hd \
  -H "Authorization: Bearer roasos_sk_live_..." \
  -H "Idempotency-Key: 0f1c9e2a-minimax-speech-hd-0001" \
  -H "Content-Type: application/json" \
  -d '{"quote_id":"q_7Hk...","text":"Two drops, morning and night. That is the whole routine."}'
# 201 { "data": { "id": "gen_...", "status": "queued", "credits_reserved": 2 } }

# 3. Poll, then read the result.
curl https://api.roasos.com/v1/jobs/gen_.../status -H "Authorization: Bearer roasos_sk_live_..."
curl https://api.roasos.com/v1/jobs/gen_.../result -H "Authorization: Bearer roasos_sk_live_..."

The deprecated path

POST /v1/generations still reaches this model for one minor version. It is deprecated. This is how it maps:

POST /v1/audio/minimax-speech-hd is the endpoint. POST /v1/generations still works with { "type": "audio", "model": "minimax-speech-hd", "prompt": "<text>", "options": { "voice": "English_expressive_narrator", "speed": 1.05, "pitch": 0, "emotion": "happy", "english_normalization": true, "language": "fr" } }; there the script goes in prompt. voice_id names a saved clone from POST /v1/voices on both surfaces. options.stability must be 0.5 and options.similarity_boost must be 0.75 (no-op).

See also

• llms.txt for this model — the same facts as one plain-text file for an agent.
• GET https://api.roasos.com/v1/catalog/minimax-speech-hd — the machine-readable form: enums, defaults, limits and prices. Refer to the API reference.
• Jobs — how to poll a job and read its result.
• Provider endpoints: fal-ai/minimax/speech-2.8-hd.