How the 12 Tools Fit Together
A practical walkthrough of what each stage does and how to move a song from a text prompt to a finished, licensed master — all inside one library.
Which tool should I start with in the Music Factory?
Start with Generate Music if you are building a song from an idea: describe the genre, mood, vocal, and instruments and the AI returns two full-song variants for 30 credits. Start with Upload & Extend or Add Vocals instead if you already have your own audio — a beat, a loop, or an instrumental — that you want to continue or sing over.
The three “Create” tools all give you a fresh track to work from. Generate Music writes from a text prompt; Upload & Extend continues an audio file you supply; Upload & Cover reinterprets any song you upload as a new AI version. From there, the “Transform” tools reshape what you have — extending, covering, adding vocals or a full instrumental band, replacing a single section, or turning a track into a music video. The “Analyze” tools pull things out of a finished track: separated stems, a reusable voice persona, or synced LRC lyrics. Because every tool reads from and writes back to the same library, you rarely need to leave the studio between stages.
What does a full song-making workflow look like end to end?
A common path is: generate a core track, add or fix the vocal, extend it to full length, polish any weak section, and then pull out lyrics or stems for delivery. Each step is its own 30-credit generation (stems 50, video 100), and each builds on the version before it, so your corrections stack rather than resetting.
In practice that means you might generate an upbeat pop idea with Generate Music, decide the chorus needs a stronger hook and rewrite just that window with Replace Section, then continue the arrangement past its natural end with Extend Track so it fits a 90-second video edit. If you started from an instrumental instead, you would upload it to Add Vocals to have an AI singer perform your lyrics over it, then treat the result as a normal track you can extend or edit. For a lyric video, you finish by running the vocalled track through Timestamped Lyrics to get an LRC file with a timestamp on every line. None of these steps require exporting and re-importing between apps, which is what keeps the key and tempo consistent across the whole project.
New to AI music? The Music Factory studio guide walks through a full session, and the online song-creation guide covers prompt-writing in more depth.
How is Music Factory different from the standalone Music Generator?
The standalone Music Generator is a focused tool for creating songs from a text prompt. Music Factory includes everything the Music Generator does and adds eleven more tools around it — extending, covering, adding vocals and instrumentals, replacing sections, making videos, separating stems, extracting lyrics, and building voice personas — so the whole production lives in one place instead of across several apps.
| Capability | Music Generator | Music Factory |
|---|
| Generate from text prompt | Yes | Yes |
| Extend / continue a track | No | Yes |
| Add vocals to an instrumental | No | Yes |
| Replace one section of a song | No | Yes |
| Separate vocal / instrumental stems | No | Yes (50 credits) |
| Generate a music video | No | Yes (100 credits) |
| Timestamped LRC lyrics | No | Yes |
Which music engine version should I use?
Music Factory offers five engine generations — V3.5, V4, V4.5, V4.5+, and V5. For most work the newest engine (V5) is the best default: it produces the clearest vocals and the most coherent arrangements. Older engines are still useful when you want a rougher, more vintage character, and a couple of features are restricted to the premium engines.
Specifically, Add Vocals and Add Instrumental run only on V4.5+ and V5, because performing a vocal over an uploaded instrumental — or building a full band under an uploaded vocal — is harder than generating a song from scratch and needs the premium models to keep the result in key and in time. Basic tools like Generate, Extend, Cover, and Replace Section work across all five engine versions, so you can match the engine to the character you are after or stay on the version a track was originally made with for the most consistent result.
Can I work with my own audio, not just AI-generated tracks?
Yes. Three tools accept files you upload rather than requiring a track that was generated in the studio. Upload & Extend continues an audio file you supply — a voice memo, a guitar loop, a half-finished demo — growing it into a longer piece from a point you choose. Upload & Cover reinterprets any song you upload as a new AI version with fresh vocals and arrangement. And Add Vocals takes an instrumental you upload and performs a sung topline over it.
The remaining tools work from tracks already in your library, because they need the underlying generation data — Extend Track, Create Cover, Replace Section, Music Video, Separate Vocals, and Timestamped Lyrics all read a track you generated earlier. In practice this means the fastest way to unlock the whole suite for your own material is to bring it in through an upload tool first: once your audio has produced a track in the library, you can then extend it, edit sections, or pull stems from it like anything else. One caveat worth repeating — uploading a copyrighted recording and transforming it does not grant you rights to that original source, even though the AI-generated output itself is licensed for commercial use.
Can I use Music Factory songs commercially, and who owns them?
Yes. Every output from Music Factory — generated songs, covers, vocalled tracks, extended tracks, and music videos — carries a full commercial license, so you can publish and monetise it on YouTube, TikTok, Spotify, in ads, games, podcasts, streams, or paid client work. Because the audio is original AI-generated content rather than lifted from an existing recording, there are no Content ID claims to clear.
The one thing to keep in mind is that the license covers the material the tool generates. If you use an upload-based tool — Upload & Extend, Upload & Cover, or Add Vocals — and the audio you upload contains someone else’s copyrighted recording, the AI output does not grant you rights to that original source. Credits are consumed at the moment you start each job, and there is no subscription: you pay per generation and keep whatever you make.
