Every question, answered.

MixGenie is an AI song generator. You describe a vibe, mood, or story in one line; it writes the lyrics, generates the vocals, and arranges the music. You get a downloadable MP3 in minutes.

Yes — every account gets free credits on signup and monthly refresh. Paid credit packs and subscription plans unlock higher volume; you never lose access to the free credits.

No. If you can describe a song in one sentence — a mood, a genre, a story — MixGenie can generate it. The lyric generator, reference-audio upload, and song-from-image flows are shortcuts for when words alone are hard.

Web at app.mixgenie.ai; installs as a PWA on iOS and Android (Add to Home Screen); a native Android app is in beta. Everything syncs through your MixGenie account.

First render is usually 2–4 minutes on the free tier; paid credits go through a priority queue with lower wait times during peak load.

Typically 2–3 minutes — full arrangement with intro, verses, chorus, bridge, outro. Not a loop.

Everything the underlying music model supports — pop, rock, electronic, hip-hop, classical, ambient, folk, jazz, cinematic scores. If your prompt names a style it recognises, it renders it.

No. Vocal generation is prompt-driven; we deliberately do not offer voice cloning. Prompts inspired by an artist produce original vocals in a similar style, not the artist's actual voice.

Yes. Edit inline in the app, then regenerate the vocals on top of the same arrangement — free, no extra credit.

Yes — paste them into the lyrics field before generating.

English, Spanish, Portuguese, French, and a few more. The vocal generator sings in the language you drafted in.

No. The reference informs style, mood, tempo, and arrangement shape — the melodies, lyrics, and vocals are freshly generated. Inspiration, not sampling.

MP3, WAV, and M4A. Files up to a few MB — the clip picker trims them locally before upload.

For personal reference generation, yes — the output is original and not a derivative work. For commercial releases, follow the same due diligence you would for any inspired composition.

One credit = one song generation. Free credits refresh monthly. Credit packs (never expire) and monthly subscriptions (larger monthly refresh + priority queue) are optional.

Free-tier credits reset monthly. Paid credit packs never expire.

Yes, one click in the app. You keep subscription features until the end of the current billing period.

Yes — you own the output of every song you generate, subject to the terms of service. No royalty split, no attribution requirement.

Yes, per the terms. See /terms.

Yes. The vocal generator returns per-word timestamps; the video renderer paints each word based on the audio clock, not a fixed offset.

9:16 (Reels/TikTok/Shorts), 16:9 (YouTube), and 1:1 (Instagram grid).

Yes — upload a custom cover from your device or camera; it replaces the AI cover in the video.

WebM by default (accepted by TikTok/Reels/Shorts/YouTube). MP4 export is in progress.

No. Your prompts, references, and generated songs are not used to train models. See /privacy.

The reference is used to generate the song and is not retained for training. Your generated songs are stored so you can access them in your library.

AWS eu-west-1 (Ireland). See /privacy for details on subprocessors.

Yes, from account settings in the app. Deletion removes your account, songs, and covers. Non-personal aggregates (usage counts) may be retained.