Documentation

Audio & Music

POST /api/v1/tasks — generate music asynchronously.

POST/api/v1/tasksBearer sk-app-…

Music generation is async. Suno-class models return a task_id; on success the result carries one or more audio URLs and (where supported) lyrics + cover art. Multi-step flows like Suno extend / cover are also reachable through dedicated model ids.

Pick an audio model#

Audio covers very different jobs — Suno generates full songs in a multi-step workflow, ElevenLabs handles TTS and short music clips, Runway Act-Two drives a character video by a reference performance.

Model ID	Model	Spec	Top params	Doc
`suno-v5`	Suno v5 Suno v5 music generation. Multi-step workflow: generate → extend / cover / lyrics / add-vocals / instrumental / mashup / stem / mv / wav / cover-img / replace / persona / upload-extend.	—	`promptlyricsinstrumental`	View →
`elevenlabs-v3`	ElevenLabs v3 (TTS) ElevenLabs v3 text-to-speech.	—	`textvoice_idmodel_id`	View →
`elevenlabs-music`	ElevenLabs Music ElevenLabs music generation.	—	`promptduration`	View →
`runway-act-two`	Runway Act-Two Runway Act-Two — drive a character video by a performance reference (audio + driver video).	—	`character_video_urldriver_video_url`	View →

4 models

Request body (shared)#

Field	Type	Default	Description
`model`required	`string`	—	Model ID, e.g. "suno-v5", "suno-extend".
`prompt`required	`string`	—	Style / mood description.
`lyrics`	`string`	—	Optional lyrics to set to music.
`duration`	`int`	`60`	Seconds, 30–240 depending on model.
`instrumental`	`boolean`	`false`	No vocals if true.
`webhook_url`	`string`	—	POST target on completion.

Response#

Field	Type	Default	Description
`task_id`	`string`	—	Use with GET /api/v1/tasks/{task_id}.
`status`	`string`	—	QUEUED on submit; RUNNING / SUCCESS / FAILED / EXPIRED later.

Suno v5 multi-step workflow#

Each Suno call returns a task_id. Chain follow-up actions by re-POSTing /api/v1/tasks with action=<sub-action> and parent_task_id set. Orux AI preserves the chain so the cost is attributed correctly and the resulting clip can be cross-referenced.

action	Description
`generate`	Initial song from prompt (+ optional lyrics).
`extend`	Continue an existing song by N seconds.
`cover`	Sing the same lyrics in a different style.
`lyrics`	Get just the LRC-format lyrics for a generated song.
`add-vocals`	Add vocals to an instrumental track.
`instrumental`	Strip vocals, keep only instrumental.
`mashup`	Combine sections of two prior tasks.
`stem`	Separate into individual instrument stems.
`mv`	Generate an accompanying music video.
`wav`	Convert the mp3 output to a high-quality WAV.
`cover-img`	Generate an album cover image.
`replace`	Replace a section of a song with a new arrangement.
`persona`	Reuse a saved Suno persona as the singer.
`upload-extend`	Upload your own audio file and extend it.

Suno v5`suno-v5`

Suno v5 music generation. Multi-step workflow: generate → extend / cover / lyrics / add-vocals / instrumental / mashup / stem / mv / wav / cover-img / replace / persona / upload-extend.

async task

•Each Suno call returns a task_id. Chain sub-actions by sending action=extend (or cover/stem/wav etc.) with parent_task_id set.

Field	Type	Default	Description
`prompt`required	`string`	—	Text prompt describing the desired output.
`lyrics`	`string`	—	Optional lyrics to sing.
`instrumental`	`boolean`	`false`	No vocals if true.
`duration`	`int (30–240)`	`60`	Length in seconds.
`action`	`enum`	`generate`	Suno sub-action. Most actions take a previous task_id as parent_task_id. `generateextendcoverlyricsadd-vocalsinstrumentalmashupstemmvwavcover-imgreplacepersonaupload-extend`
`parent_task_id`	`string`	—	Reference to a previous Suno task when chaining actions like extend or cover.
`webhook_url`	`url`	—	HTTPS endpoint Orux AI will POST a signed event to on terminal status. Optional — fall back to polling.

parent_task_id glues the chain together

Every sub-action passes the prior task_id as parent_task_id. The dashboard groups all derived tasks under the original Suno song so cost attribution stays sane.

ElevenLabs v3 (text-to-speech)#

Synchronous TTS via ElevenLabs v3. Pass text and voice_id; the optional stability / style / use_speaker_boost knobs trade off naturalness vs. expressivity. The voice_id catalogue is the same as on ElevenLabs.

ElevenLabs v3 (TTS)`elevenlabs-v3`

ElevenLabs v3 text-to-speech.

async task

Field	Type	Default	Description
`text`required	`string`	—	Text to read aloud.
`voice_id`required	`string`	—	ElevenLabs voice id.
`model_id`	`string`	`eleven_v3`	Optional model override.
`stability`	`float (0–1)`	`0.5`	Voice stability 0–1.
`style`	`float (0–1)`	`0.3`	Style exaggeration 0–1.
`use_speaker_boost`	`boolean`	`true`	Boost speaker similarity.
`webhook_url`	`url`	—	HTTPS endpoint Orux AI will POST a signed event to on terminal status. Optional — fall back to polling.

ElevenLabs Music#

Short-form music generation, 10–120 seconds. Lower-latency alternative to Suno for sound effects, jingles and game audio.

ElevenLabs Music`elevenlabs-music`

ElevenLabs music generation.

async task

Field	Type	Default	Description
`prompt`required	`string`	—	Text prompt describing the desired output.
`duration`	`int (10–120)`	`30`	Length in seconds.
`webhook_url`	`url`	—	HTTPS endpoint Orux AI will POST a signed event to on terminal status. Optional — fall back to polling.

Runway Act-Two#

Drive an existing character video to mimic a performance reference. Pass character_video_url (the actor you want to animate) and driver_video_url (the source performance). Suited for retargeting motion, dance and dialogue without re-shooting.

Runway Act-Two`runway-act-two`

Runway Act-Two — drive a character video by a performance reference (audio + driver video).

async task

Field	Type	Default	Description
`character_video_url`required	`url`	—	The character to be animated.
`driver_video_url`required	`url`	—	The performance video to mimic.
`webhook_url`	`url`	—	HTTPS endpoint Orux AI will POST a signed event to on terminal status. Optional — fall back to polling.

Examples#

Submit a song

curlshell

curl https://orux.top/api/v1/tasks \
  -H "Authorization: Bearer $ORUX_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "suno-v5",
    "prompt": "Lo-fi hip hop beat with rainy night ambience",
    "lyrics": "Soft rain on neon glass, the city hums...",
    "duration": 60
  }'
# -> { "task_id": "task_01HZY...", "status": "queued" }

Extend a previous Suno song

curlshell

# Continue a previous Suno song by 30 seconds
curl https://orux.top/api/v1/tasks \
  -H "Authorization: Bearer $ORUX_API_KEY" \
  -d '{
    "model":"suno-v5",
    "action":"extend",
    "parent_task_id":"suno_01HZY...",
    "duration": 30,
    "prompt":"keep the dreamy synth lead, add subtle drums"
  }'

ElevenLabs v3 text-to-speech

curlshell

curl https://orux.top/api/v1/tasks \
  -H "Authorization: Bearer $ORUX_API_KEY" \
  -d '{
    "model":"elevenlabs-v3",
    "text":"Welcome to Orux AI. One key, every model.",
    "voice_id":"EXAVITQu4vr4xnSDxMaL",
    "stability": 0.5,
    "style": 0.3,
    "use_speaker_boost": true
  }'