Mirelo SFX 2.0 (Text-to-Audio)

Mirelo SFX 2.0 makes sound effects and ambience from a text description, from one second to ten minutes long, in up to four variations.

Mirelo SFX 2.0 is Mirelo's sound-effect model, released in October 2026. Describe a sound and it generates it: a door creaking open, rain on a metal roof, footsteps on gravel, a cheering stadium or a quiet ambience bed for a scene.

Say what the sound should be in the prompt, up to 2000 characters, and what to keep out in negative_prompt, for example music or voices. duration sets the length in whole seconds, from 1 to 600; num_samples asks for one to four variations of the same prompt; seed repeats a result; upload_audio_format picks WAV (the default), MP3, AAC or FLAC.

Inputs: prompt (required), negative_prompt, duration (default 10 seconds), num_samples (default 1), seed and upload_audio_format. Output: audio, one file per variation.

Price: $0.015 per second of sound for each variation: $0.15 for the default 10 seconds, $0.90 for two 30-second variations, at most $36.00 for four of 10 minutes. It makes sound effects and ambience from text alone: it does not watch a video, and it does not make music or speech.

Price: $0.015 per second of duration. One run with default settings costs $0.15.

Run Mirelo SFX 2.0 (Text-to-Audio) via API

curl -X POST https://api.flymy.ai/api/v1/flymyai/mirelo-sfx-2_0-text-to-audio/predict \
  -H "X-API-KEY: $FLYMYAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt":"Heavy rain on a metal roof with distant thunder","duration":10}'

API documentation · Create a free account · Pricing

Related models