AI Music Video Generator by Dimly AI

Dimly AI is an AI music video generator and virtual cinematographer. Bring a Suno track or your own song and Dimly plans connected footage, aligns available lyrics and delivers a finished music video with your source audio. It also runs an AI Rap Duo Generator: two photos become a 12-second video of both people rapping an original song written for them.

Yes. Attach photos in the chat and the director builds character sheets from them that keep the face, then uses those sheets for every shot the person appears in. Use only photos of people who have agreed to it.

Yes, two ways. Lyrics can be written into the picture itself — on a sign, in light, on a wall — or added as captions over the finished cut. The director asks which you want; you can also have neither.

Yes, planning is free, and so are your first 6 images on an account that has never paid. Create an account (no card needed), upload a song or paste a Suno link, and Dimly analyzes it (sections, beats, lyric timing), plans the whole video with you and makes your first reference sheets and keyframes at no cost. Video always uses credits, and you confirm a quote before anything renders. Plans start with the Weekly Pass, and one-time credit packs need no subscription.

Yes. A full song uses more credits than a chorus cut, so the plan matters: Pro and Creator cover a 3-minute song, Starter covers a shorter one, and the Weekly Pass is sized for chorus cuts. The Song and Studio credit packs also cover one. The pricing page lists what a 3-minute song costs on each model, and the quote shows the exact number before anything renders.

Write to the director in your own language and it answers in that language. Your lyrics stay in their original language on screen and in captions.

Each planning reply takes about a minute while the director reads the song and works out the next step. Once you approve the quote, clips render in the background; a full song takes longer than a chorus, and you can come back to the project to review takes as they finish.

Yes. Tell it to make the choices and it will decide the look, the cast and the storyboard on its own. It still stops once before spending: you approve the quote, and only then does anything render.

Yes. Paste a public Suno track link or upload your own MP3, WAV or M4A. Song analysis runs first — waveform, structure, lyric timing — so the director works from the song itself. The full song, a lyric video, a visualizer and Chorus Cut all start from this same step.

Generic tools generate disconnected clips from text prompts. Dimly is an AI video agent: a director that plans a connected sequence around your song, keeps faces, light and camera consistent across every shot, and brings you concrete choices instead of waiting for better prompts.

Each shot is routed through vetted video models, chosen for lighting, texture and continuity — and the routing can change as newer models pass review. It never changes your quote: every section shows its cost before it renders.

You can publish work when you have the rights to the source audio and inputs. Generated output is also subject to applicable provider and plan terms, so review them for commercial campaigns.

If you own the rights or hold permission for the source song and footage, you can use the finished video commercially. Generated output also follows your plan’s terms. Free exports and the smallest credit pack carry the Dimly watermark; any plan, and the Song and Studio packs, export without it. Review both before a client campaign.

Not for a music video. You bring the music; Dimly decides how the whole song is seen — style, direction, the shots and the cut. Only here for the chorus? Chorus Cut runs the same flow and closes it in a 15-second cut. Captions come from lyric timing on your own audio; nothing invents lyrics for a track you did not provide. The one exception is the AI Rap Duo Generator: there Dimly writes and records an original 12-second rap from two photos and the occasion you describe.