Generate expressive AI speech with Suno Speech. Control tone, emotion, vocal gender and background music. Pay per generation, no subscription. Generate expressive AI speech with tone, rhythm, emotion and optional background music from a scene prompt or a full script.
Generate expressive AI speech with tone, rhythm, emotion and optional background music from a scene prompt or a full script.
About this model
Suno Speech generates spoken audio with natural tone, rhythm and emotion, and can optionally mix in background music. It is not a fixed-voice text-to-speech engine: describe the scene and the model writes and performs the script, or supply your own script and steer the delivery with tone, variety, vocal gender and background music controls. It is a good fit for narration, news-style intros, character lines, ads and social voiceovers where performance matters as much as the words.
Simple mode takes a one-line scene prompt (up to 3000 characters). Advanced mode takes an exact script (up to 5000 characters) plus optional tone instructions of up to 1000 characters. Pair it with Suno Create Music for full songs or Suno Sounds for sound effects.
Cost analysis
| Provider | Cost | Notes |
|---|---|---|
| muapiapp | $0.10 per generation | Same price as Suno Create Music, pay per generation. |
| Fal.ai | Not available | No Suno Speech endpoint confirmed. |
| Replicate | Not available | No Suno Speech endpoint confirmed. |
Same price as Suno Create Music, pay per generation.
No Suno Speech endpoint confirmed.
No Suno Speech endpoint confirmed.
** Competitor pricing is estimated based on similar model architectures and usage tiers.
Configuration schema
| Parameter | Type | Description | Default |
|---|---|---|---|
| Custom Mode | boolean | Off (Simple): describe a scene in `prompt` and the model writes and speaks the script. On (Advanced): provide your own `script` plus optional tone, variety, vocal gender and background music. | false |
| Scene Prompt | string | Simple mode only. A one-line scene description; the model writes the script. 1-3000 characters. | A short morning news intro about sunny weather |
| Script | string | Advanced mode only. The exact text to be spoken, 1-5000 characters. Multiline supported. | Welcome to today's briefing. The weather is clear and the city is awake. |
| Tone | string | Advanced mode only, optional. Tone, rhythm, emotion, setting, accent or accompaniment style. Up to 1000 characters. | Warm, calm female narrator, slightly smiling |
| Variety | Enum (5 options) | Advanced mode only, optional. 0 = exact style, 1 = balanced variety, 2 = distinct styles, 3 = bold exploration, 4 = unreasonably varied. | - |
| Vocal Gender | Enum (2 options) | Advanced mode only, optional. Voice gender. | - |
| Background Music | boolean | Advanced mode only, optional. Add background music under the speech. | - |
Off (Simple): describe a scene in `prompt` and the model writes and speaks the script. On (Advanced): provide your own `script` plus optional tone, variety, vocal gender and background music.
falseSimple mode only. A one-line scene description; the model writes the script. 1-3000 characters.
A short morning news intro about sunny weatherAdvanced mode only. The exact text to be spoken, 1-5000 characters. Multiline supported.
Welcome to today's briefing. The weather is clear and the city is awake.Advanced mode only, optional. Tone, rhythm, emotion, setting, accent or accompaniment style. Up to 1000 characters.
Warm, calm female narrator, slightly smilingAdvanced mode only, optional. 0 = exact style, 1 = balanced variety, 2 = distinct styles, 3 = bold exploration, 4 = unreasonably varied.
-Advanced mode only, optional. Voice gender.
-Advanced mode only, optional. Add background music under the speech.
-Developer documentation
Pick a mode: Leave custom_mode off (Simple) and describe a scene in prompt, or turn it on (Advanced) and provide your own script.
Steer the delivery (Advanced only): Use tone for emotion, rhythm, accent and setting, variety (0-4) for how far the model may stray from the described style, vocal_gender (male or female), and background_music to add music under the voice.
Submit and fetch the result: The call returns a request_id; poll the result endpoint or pass a webhook.
curl -X POST https://api.muapi.ai/api/v1/suno-speech \
-H "Content-Type: application/json" \
-H "x-api-key: $MUAPI_KEY" \
-d '{
"custom_mode": true,
"script": "Welcome to today'\''s briefing. The weather is clear and the city is awake.",
"tone": "Warm, calm female narrator, slightly smiling",
"vocal_gender": "female",
"background_music": true
}'
Simple mode example: {"custom_mode": false, "prompt": "A short morning news intro about sunny weather"}.
Frequently asked
Suno Speech generates spoken audio with controllable tone, rhythm and emotion, with optional background music. Unlike fixed-voice TTS, the performance is shaped by your prompt or tone description.
Simple mode (`custom_mode: false`) takes a short scene `prompt` and the model writes the script. Advanced mode (`custom_mode: true`) speaks your exact `script` and accepts optional `tone`, `variety`, `vocal_gender` and `background_music`.
Prompts can be up to 3000 characters in Simple mode. Scripts can be up to 5000 characters in Advanced mode, and tone descriptions up to 1000 characters.
`variety` is an integer from 0 to 4: 0 follows the described style exactly, 1 is balanced, 2 gives distinct styles, 3 is bold exploration and 4 is unreasonably varied. It only applies in Advanced mode.
Yes. Set `background_music` to true in Advanced mode and the speech is mixed with a music bed. You can also describe accompaniment style in `tone`.
Each generation costs the same as a Suno Create Music request: $0.10 per request, billed only when you generate.