Music GenerationActive

ElevenLabs Text to Music v2.5

Create original music tracks from one written brief using ElevenLabs Music v2.5. Set track length and MP3 quality, and force instrumental-only output when needed.

Text to Music

Model ID

text to music v2.5

Provider

elevenlabs

Updated

Nov 4, 2026, 13:10

Support Wiro on G2

wiro playground—elevenlabs/text to music v2.5

Updated Nov 4, 2026, 13:10

Overview

Text to Music v2.5 is an audio generation model from ElevenLabs. You describe the song in plain language, then the model composes and produces a finished music track as an MP3 file. It’s useful when you need custom background music that fits a scene, brand mood, or gameplay loop without licensing a stock track.

What you can build

Background music for product videos, trailers, and social clips
Loopable beds for podcasts and livestreams (intro/outro cues)
Game ambience and level themes (calm, tense, combat, victory)
Draft demos for songwriting and arrangement ideas
Instrumental practice tracks in a specific style and tempo

Inputs

A written music brief that names genre, mood, instruments, tempo (BPM), and structure. You can include sections like intro, verse, chorus, and outro.
A target track length, chosen in seconds, from 10 to 300 seconds.
An MP3 quality choice that sets the sample rate and bitrate. Options include 22.05 kHz at 32 kbps, or 44.1 kHz at 32, 64, 96, or 128 kbps.
An optional instrumental-only switch. Turn it on to block vocals. Turn it off to let the brief decide.

Outputs

One MP3 audio file that contains a complete mixed track.
The audio length matches the requested duration.
The file’s sample rate and bitrate match the quality option you picked, which affects size and clarity.

Limitations

This model run is designed for prompt-based music generation. It does not take an audio reference track or an edit range for inpainting.
The model can miss very specific musical requests, like exact chord voicings or bar-perfect hits.
If you don’t force instrumental-only output, vocals may appear when your brief suggests a sung song.
Low-quality prompts cause unstable results. Vague notes like “make it good” often produce generic tracks.
Requests that include copyrighted lyrics, or ask for a song “in the style of” a named artist, may be rejected.

Safety & compliance

Don’t include copyrighted lyrics in your brief.
Don’t ask for direct imitation of a specific artist, band, or song.
Your rights to use outputs can depend on the terms tied to your ElevenLabs Music access and plan.

Example prompts

Great starting points for text to music v2.5.

Smoky late-night jazz trio with brushed drums, upright bass walking lines and a warm Rhodes piano solo, relaxed 90 bpm swing, intimate club atmosphereMusic Generation

Smooth modern R&B with deep sub bass, crisp finger snaps and lush chords, a soulful female vocal singing about city lights at midnight, 85 bpm, polished productionMusic Generation

Euphoric melodic house with a driving four-on-the-floor kick, rolling bassline, shimmering arpeggiated synths and a big uplifting breakdown, 124 bpm, festival energyMusic Generation

Upbeat country rock with twangy electric guitars, steady stomping drums and a gritty male vocal singing about a road trip through the desert, 110 bpm, sing-along chorusMusic Generation

API quick start

Run text to music v2.5 with a single API call.

POST https://api.wiro.ai/v1/Run/elevenlabs/text-to-music-v2-5

{
  "prompt": "Smoky late-night jazz trio with brushed d…",
  "duration": 30,
  "outputFormat": "mp3_44100_128",
  "instrumental": "true"
}

curl

curl -X POST "https://api.wiro.ai/v1/Run/elevenlabs/text-to-music-v2-5" \
  -H "Content-Type: application/json" \
  -H "x-api-key: YOUR_WIRO_API_KEY" \
  --data-binary @- <<'JSON'
{
  "prompt": "Smoky late-night jazz trio with brushed d…",
  "duration": 30,
  "outputFormat": "mp3_44100_128",
  "instrumental": "true"
}
JSON