Audio Models - Text to Sound Effects (MOSS)

What it does

MOSS-SoundEffect v2.0 takes one English sentence describing a sound and returns that sound as a 48 kHz mono WAV, from half a second to 30 seconds long.

It is built for effects and ambience — impacts, foley, animals, weather, interface sounds, transitions — rather than music or speech. Every clip comes out at the same peak level, so you set the balance against narration and music in Audio Mix rather than fixing each clip's level first.

To lay several effects on one timeline, generate each one, gather them with Sound Collect, and feed the list to Audio Mix's Effects input.

Problem it solves

• Foley for short video — Footsteps, a door, keyboard typing, a glass breaking
• Ambience beds — Rain, wind in trees, birds in a park, crowd applause
• Interface and transition sounds — A notification ding, a camera shutter, a whoosh
• Cartoon and magic accents — A boing, a sparkle chime
• Multi-effect timelines — Several effects placed on one timeline with Sound Collect and Audio Mix

Input/Output

How to Use

1. Write a focused English description in a Text Input — one sound per prompt.
2. Connect it to MOSS and set Duration to the length of the event.
3. Send the output to Sound Output, run, and audition. Revise the description if the source or texture is wrong.
4. Trim or adjust the clip with Audio Edit if needed.
5. For several effects, gather the clips with Sound Collect and feed the list to Audio Mix's Effects input; use Video Hit Points to find where each one should land.

Example Workflow

Prompt: "A short, soft metallic click as a small latch closes, dry close-up recording."

Duration: 1 second

Flow: Text Input → MOSS → Sound Output.

For a full short-form sound pass, see the Shorts Sound Effects recipe.

Use Case Tips

• Describe the audible event — Name the source, the action, the texture, and the space ("footsteps on a wooden floor in a small room"), not the whole visual scene.
• One sound per prompt — Generate a door and footsteps separately, then combine them in Audio Mix.
• Duration is free — Generation takes about the same time for 1 second as for 30 (the model always renders the full 30-second window and trims it), so set the length you actually need.
• Balance in the mix — Clips are peak-normalized; lower effects under dialogue in Audio Mix.

Limitations

• Prompts are English. Other languages are not supported by this node.
• Output is mono. For stereo width, place effects in Audio Mix or use a stereo source.
• Not for speech or music — use a TTS node or Text to Music instead.
• Generated audio is synthetic and must be marked as such under the EU AI Act.

How it compares

Choose MOSS for effects and ambience, Text to Speech (Qwen3 / Qwen3 Voice Design / Gemini) for narration, and ACE-Step 1.5 or Lyria for music.

Model Source

• CNAPS model page: MOSS-SoundEffect v2.0
• Hugging Face: OpenMOSS-Team/MOSS-SoundEffect-v2.0
• GitHub: OpenMOSS/MOSS-TTS

Compliance & Provenance

For platform classification, acceptable use, and output marking, see the CNAPS AI Policy.

Run it now

Try it in the Shorts Sound Effects recipe, or open CNAPS Studio, find MOSS in the Component panel, and start with the example workflow above.