A music clip is a short excerpt — the hook, not the whole record. This platform generates full tracks rather than clip-length files, so the honest workflow is: generate, choose the section you want, and cut it. This page covers each step and the limits you should plan around.
- Generate a track in the AI Music Generator. Sign in first: an account is required, and each task consumes credits at the cost shown for the selected model.
- Make sure the hook arrives early. In the prompt, describe an arrangement that states its main idea in the first section rather than saving it for a later build — a clip has no time to wait.
- Shape the right section. The Extend tool lets you continue a track from a chosen point in seconds. Use it when you want a longer version of the part you intend to cut, or a cleaner ending for the excerpt.
- Download the track from the result panel.
- Cut the clip in your editor. Trim to 15–30 seconds, land the cut on a bar line, and fade in and out over a beat or two so the excerpt does not start or end mid-phrase.
- Check the destination's rules. Short-form platforms have their own policies for AI-generated audio and for using commercial music in uploads — read them before you publish.
- A teaser for a track you are releasing later.
- Vertical social edits where 15–30 seconds is the whole piece.
- A background bed for a product clip or a slide sequence.
- Ringtones and short notification-style loops, provided the platform you use allows them.
- Judge the first four bars. If the opening does not grab attention on its own, no clip will fix that.
- Prefer a change. A section transition gives the clip a shape; a flat loop does not.
- Keep vocals out of the cut-in and cut-out. Trim on instrumental bars if the model gave you any.
- Use Instrumental Mode when the clip sits under speech — see the AI Instrumental Generator page for how to run it.
- There is no clip-length output setting. Length belongs to the model's task, so you cut the excerpt yourself.
- Clips are not beat-aligned for export. There is no loop-point or grid export; align your cut by ear in the editor.
- No per-instrument stem by default. Multiple-stem separation from the workbench can give you parts after the fact, but it is a separate billed task.
- No automatic commercial rights. Rights are not automatically granted and depend on the selected model provider, your plan, your inputs, applicable law and third-party rights — check the Terms of Service.
- Clip use is still your responsibility. Some platforms restrict AI audio in their libraries; confirm before uploading.
- Cutting before the structure is settled. Lock the track first, then cut once.
- Trimming mid-word. If a vocal straddles the boundary, nudge the cut to the bar or the phrase break.
- Re-exporting in the wrong loudness range. Normalise after the cut so the clip matches the rest of your uploads.
- Assuming the clip inherits release clearance. A short clip is still the same audio file legally; the rights question does not shrink with the duration.
No. The generator produces tracks, and clip length is something you set when you cut the excerpt. If you need a specific duration, describe a short structure and trim to it.
Use Extend with the source track, set the continue-from point in seconds, and describe the next section. That gives you more material to choose the clip from.
Technically yes, once you cut and export it. Before you distribute it, check the rights position for the model you used and the rules of the store or platform you are uploading to.
Yes. Extend is a task like any other, billed at the cost shown before you submit. Current packs and plans are on Pricing.
Open the AI Music Generator →
