Documentation

BlaqueSoul Studio is an AI music studio built specifically around Afrobeats, Amapiano and Afro Gospel — fifteen regional African sounds in all, each carrying its own tempo, instruments and arrangement conventions. Describe what you want in plain language and get a finished, playable track with vocals in about a minute.

You don't need to play an instrument or know how to produce. If you do produce, the stems and WAV exports drop straight into your DAW.

Simple and Custom

Simple is one box: describe the song and generate. The AI writes its own lyrics, picks its own structure and titles the track.

Custom adds a track title, the AI Lyrics Composer, your own lyrics, and Advanced options — things to avoid, three creative-control sliders (style adherence, weirdness, audio influence) and track length on the newest model.

Sound profiles

Fifteen regional profiles, each carrying the tempo band, signature instruments and arrangement conventions of a specific sound — so the platform's knowledge of your genre is visible rather than something you have to trust. Pick one from the style menu on Create and it shapes the whole generation.

The AI Lyrics Composer

Built for African music specifically, not a generic lyric generator with an African setting bolted on. Pick your options, generate, and edit the result freely. 3 credits per generation.

• Language — English, Pidgin, Swahili, Amharic, Shona, Ndebele, Zulu, Xhosa, Twi, Yoruba, Igbo, Hausa, Wolof, Bambara, Krio, Lingala, Portuguese, French, Arabic, or any language you type. Written natively, not translated.
• Genre, mood and theme — from Afrobeats to Maskandi; heartbreak to celebration.
• Artist gender — also sets the vocal for the generation.
• Structure — full verse/chorus, short, verse/hook or freestyle.
• Explicit content — clean or mature.

Personas

Reusable artist identities — a name, a description and a vocal character — so a series of tracks can sound like the same artist. Create one from scratch or from a track you've already made.

Cloning your own voice

Voice cloning is live. Record or upload a sample, then read aloud the verification phrase the system gives you — a consent check that proves the voice is yours. Your voice then becomes selectable anywhere you'd pick a persona.

50 credits, refunded if it fails. Up to two cloned voices per account; delete one to make room. If a phrase recording doesn't pass you can regenerate the phrase and try again without paying twice.

Licensing your voice

Once a voice is ready you can list it in the Voices marketplace and set a per-use price between 1 and 200 credits. Every listing is reviewed by a human before anyone can use it.

• You keep 80% of the fee; the platform takes 20%.
• You're paid when the generation succeeds, not when the buyer is charged — so a failed job refunds the buyer in full and there is never a payment to claw back.
• Earnings appear on your dashboard and in Settings, itemised per voice.
• You can withdraw a listing at any time.

The Track Canvas

One screen per track where the waveform is the interface. Drag across the wave to select a section and rebuild just that part, stack vocals or instruments on top, extend the ending, or line the separated stems up against the lyrics.

Your original is never overwritten — every edit lands as an A/B alternative you can switch between. A replaced section must be between 6 and 60 seconds and costs the usual 36 credits, since it generates new audio. Open it from the Canvas item in the sidebar, or from any track in your library.

Everything else

• Extend — continue a song past its ending.
• Remix / fork — open a public track (or your own) as the start of a new one.
• Cover — re-perform a track in a new style or voice.
• Upload & extend — bring your own audio in and have the AI continue it.
• Add vocals / add instrumental — put a vocal over an instrumental, or build a backing track under vocals.
• Mashup — blend two of your own tracks into one.
• AI cover art — generates artwork options; pick one and apply it.
• Music video — renders a video for the track.

What it is

A multitrack timeline — the place a song goes once it needs more than a single waveform. Lay takes, stems and your own uploads out against each other, arrange them, mix them, and export a master or a full multitrack bundle.

It uses the same account, library and credits as the rest of BlaqueSoul. There is no second login and no second copy of anything. Open it from Studio in the sidebar, or start a project straight from a song or an upload.

Getting audio in

• Your songs — every take of every generation, and your covers.
• Stems — a completed split arrives as aligned lanes, in mixer order, all starting together.
• MIDI — a transcription lands as a lane you can audition.
• Uploads — drop a file straight onto the Studio and it lands on the timeline and stays in your library. MP3, WAV, FLAC, M4A or OGG, up to 100MB.

Nothing is duplicated. A clip points at the library row it came from, which is why importing is instant and free.

Arranging

• Drag clips between lanes, trim either edge, split at the playhead, duplicate.
• Slice a clip into bars, into beats, or at the edges of your selection.
• Snap to a bar, 1/2, 1/4, 1/8, 1/16 — or off.
• Tools for pointer, razor, eraser and zoom, so cutting twenty clips is twenty clicks.
• Everything is reachable from the keyboard: arrows move a clip, ⌥ trims it, ↑↓ change lane, S splits, space plays.
• Autosave, plus saved versions you can roll back to — and the restore is itself undoable.

Transport and tempo

Position in bars.beats.ticks alongside the clock, loop with left and right locators, and a metronome that follows the project tempo.

The Studio works the tempo out from the audio when a project opens, so the grid, snapping and bar numbers line up with the music rather than with a default of 120. Very fast tracks can come back at half speed — that ambiguity is real, and a listener has it too — so there are ÷2 and ×2 buttons next to the tempo, and a Detect button to run it again.

Mixing

• A channel strip per lane: fader, pan, mute, solo, and a level meter with a peak hold and a clip indicator.
• A three-band EQ on every lane and bus, with a curve you drag and a live spectrum of the lane behind it.
• Output busses — put six stems under one fader, pan and mute.
• Sends — feed a copy of a lane into a bus at its own level.
• A master fader that applies to the export exactly as it does to what you are hearing.

What you hear is what you download. Every control on the mixer is one the export applies the same way — which is also why there is no compressor yet: the browser and the renderer do not agree on its curve, and a control that changes the file but not the preview is worse than one that is missing.

AI on a section

Drag across the ruler to select a span and rebuild just that part — the same infill the Canvas uses, run on exactly the window you chose. The price and your balance are shown before you click.

The result is the whole song with that section rebuilt, so you can drop it on its own lane as an A/B, or comp it in — the rebuild trimmed to your section, and that section taken out of the original, so one take plays at a time. Undo puts both back whole. A section must be between 6 and 60 seconds and costs the usual 36 credits. Close the tab and the result is still waiting when you come back.

Exporting

• Full mixdown — every audible lane, mixed and limited, as a WAV.
• Selected range — the same, for the span you picked.
• Multitrack ZIP — every clip as its own file, MIDI lanes as real .mid files, plus an arrangement.json so another DAW can rebuild the layout.

Arranging, mixing and exporting in the Studio are free. Only the AI actions cost credits.

What it does not do yet

Written down so nothing here is a surprise:

• No effects beyond EQ — no compressor, reverb or delay yet.
• No pitch or speed controls.
• MIDI is for reference — it exports as .mid in the multitrack ZIP, and the built-in tone is for auditioning notes rather than for a master. It is not rendered into the WAV mixdown.
• Small screens fit, but this is not a phone-first DAW — the panels fold away and the timeline works, and it is still a desk job.
• A finished export is handed straight to your browser, so the export history can list a render you cannot download again. Save the file when you make it.

Status Clips

A vertical video with your lyrics burnt onto it, sized for WhatsApp status and TikTok — 1080×1920, between 5 and 30 seconds. Pick the hook, render it in about ten seconds, then share straight from your phone or download it.

Every clip links back to your public track page, so people who see it can hear the whole song. Clips are free — no credits. Make one from any finished track in your library.

Karaoke lyrics

Play any track and hit the mic button on the player to follow the lyrics word by word, highlighted in time with the vocal. Tap a line to jump to it. It works on public track pages too, so anyone you share with can sing along without an account. Free, and available on tracks that have lyrics.

Public track pages

Every track is private by default. Publish one and it gets its own shareable page that anyone can play without signing up — and it appears on the Vibe Feed. Nothing is published unless you choose to publish it.

• Vibe Feed — public tracks from across the platform, viewable without an account.
• Profiles — a public page per creator with their published tracks, bio, location, website and avatar.
• Follow other creators and see who follows you.
• Daily challenges — a rotating brief. Accept it, make a matching track, and 5 credits land automatically. One reward per challenge.

The weekly chart

Completing a challenge with a track puts it on the chart. Anyone can then vote, including people without an account looking at the public page — though voting itself needs one.

• Position is votes weighted three times a like, so deliberate support outweighs a passing tap.
• One vote per person per entry, enforced by the database — a double-tap can't inflate a chart. Change your mind and you can withdraw it.
• You can't vote for your own entry.
• Past winners are kept, so the chart has a history.