Upload a front-facing portrait, attach a song (an MP3 or WAV upload, including a song downloaded from Suno, or a YouTube or TikTok link), trim the waveform to the hook you want, pick 480p or 720p, and generate. A 10 to 30 second clip renders in a few minutes, after which you can burn timed lyric captions into it.
About five minutes end to end. Uploading the photo and song takes a minute, trimming takes another, and a 10 to 30 second clip renders in a few minutes. Pasting a YouTube or TikTok link adds roughly 90 seconds while Solmi fetches the audio server-side.
Yes, to start. A new account on the web gets 100 one-time music-video credits with no card required, enough for about 25 seconds of 480p singing video: singing videos cost 4 credits per second at 480p and 8 at 720p, so a 10-second 480p clip is 40 credits. After that, a free account's credits refill to 10 a day, about 2 seconds of 480p, so longer clips need a paid plan or a credit pack.
A clear, front-facing portrait with one visible face, even lighting, and nothing covering the mouth — no microphone, hand, hair, or hard shadow. Higher-resolution photos hold up much better at 720p, and Solmi has a built-in AI enhancement pass for soft originals.
Up to 60 seconds per clip. For a full song, render each section as its own clip and stitch them together, or hand the whole track to an AI music video generator that handles full-length songs.
Crop the photo to portrait before uploading. Solmi's singing output follows the shape of the source photo rather than an aspect setting, so a portrait crop gives you a vertical 9:16 video for TikTok, Reels, and Shorts, and a landscape crop gives you widescreen for YouTube.