Generates sound effects and ambience from a text description, up to 600 seconds long, with up to 4 variations per request.

mirelo-ai/sfx2.0/text-to-audio

Generates sound effects and ambience from a text description, up to 600 seconds long, with up to 4 variations per request.

Inference

Commercial use

Partner

Schema

Playground API

Input

Prompt*

Additional Settings

Customize your input with more control.

Introductory pricing applies October 14–December 13, 2026, inclusive. Standard pricing resumes December 14, 2026. Billing applies to total output seconds across all samples.

Result

Idle

What would you like to do next?

Introductory pricing applies October 14–December 13, 2026, inclusive. Standard pricing resumes December 14, 2026. Billing applies to total output seconds across all samples.

mirelo-ai/sfx2.0/text-to-audio

Schema

Input

Prompt*

Additional Settings

Customize your input with more control.

Result

Idle

What would you like to do next?

Parameter

What it does

prompt

What the sound effect should sound like. Up to 2,000 characters.

negative_prompt

Optional. Sounds to keep out of the result. Up to 1,000 characters.

duration

Seconds of sound to generate, from 0.1 to 600. Defaults to 30.

num_samples

1 to 4 variations per request. Defaults to 2.

seed

Seed of the first sample. -1 or unset is random.

upload_audio_format

wav (default), mp3, aac or flac.

upload_audio_bitrate_kbps

Optional bitrate for mp3 (32 to 320, in fixed steps) and aac (32 to 256) outputs. Unset uses a high-quality default.

Mirelo SFX 2.0 Text to Audio

Mirelo SFX 2.0 turns a description into a sound effect. Write what it should sound like and get back audio you can use as a one-shot, an ambience bed or a layer in a mix.

Use cases: Sound effect libraries | Ambience beds and room tone | Game and app audio | Filling gaps in a mix

How it works

Your prompt is rewritten into the caption form the model was trained on, so a plain description works: "heavy rain on a metal roof with distant thunder". The clip length is always the duration you ask for. Describe the sound, not the scene, and use negative_prompt to keep unwanted sounds out.

Every request returns one clip per sample, so you can pick the take you like best. Outputs that would clip are scaled down to a peak of -1 dBFS.

Inputs

Parameter	What it does
`prompt`	What the sound effect should sound like. Up to 2,000 characters.
`negative_prompt`	Optional. Sounds to keep out of the result. Up to 1,000 characters.
`duration`	Seconds of sound to generate, from 0.1 to 600. Defaults to 30.
`num_samples`	1 to 4 variations per request. Defaults to 2.
`seed`	Seed of the first sample. -1 or unset is random.
`upload_audio_format`	`wav` (default), `mp3`, `aac` or `flac`.
`upload_audio_bitrate_kbps`	Optional bitrate for `mp3` (32 to 320, in fixed steps) and `aac` (32 to 256) outputs. Unset uses a high-quality default.

Output

audio holds one file per sample and seeds the seed of each, in the same order. model names the release that made the output.

Pricing

Billed per second of generated audio, per sample. Two 10-second samples are 20 billed seconds.

Video to Audio generates sound that follows a video.
Video to Video returns the video with the soundtrack mixed in.

mirelo-ai/sfx2.0/text-to-audio

Input

Result

What would you like to do next?

Input

Result

What would you like to do next?

mirelo-ai/sfx2.0/text-to-audio

Input

Result

What would you like to do next?

Input

Result

What would you like to do next?

Input

Result

What would you like to do next?

Input

Result

What would you like to do next?

Input

Result

What would you like to do next?

Mirelo SFX 2.0 Text to Audio

How it works

Inputs

Output

Pricing

Related endpoints

Mirelo SFX 2.0 Text to Audio

How it works

Inputs

Output

Pricing

Related endpoints