Music WeekMauritius 2026
Day 3Tuesday · 17 June 2026

Music Production & Technology

From beat-making to AI — the tools, studios and platforms shaping how Mauritian music is made, mixed and released to the world.

Discipline
Production · Sound design
Theme
AI x Creole sound
Output
Live releases on NOUZIK
Streaming
Studio cam · 12h
Today's narrative

The story of Day 3.

If Heritage Day was about where the sound comes from, Production Day is about where it goes next. Studios open their doors, producers open their sessions, and the question on the table is honest: how do we make Mauritian music that travels — without losing what makes it ours?

The day runs in two parallel tracks. In the labs, beat-makers move from a blank DAW to a finished séga-tipik-flavoured loop in a single afternoon, with mentors looking over shoulders. In the talks, the conversation gets bigger: AI as a co-writer, not a replacement; sample ethics with traditional material; what a NOUZIK release actually looks like from upload to royalty.

By evening, the work is heard. Producers play unreleased tracks on big monitors, the resident BerTz takes the decks, and the first AI x Creole Sound showcase puts machine-assisted compositions on stage next to live percussion — same room, same audience, no apology.

Full Schedule

From morning workshops to the late-night set.

  1. 09:30

    Beat-Making Lab — Beginners

    NOUZIK Studio, Ebène

    From empty session to first loop. Bring a laptop with any DAW. Mentors rotate every 45 minutes.

  2. 11:00

    Advanced Producers Roundtable

    NOUZIK Studio, Ebène

    Mixing low-end for séga and seggae, sidechaining ravanne samples, mastering for streaming.

  3. 13:00

    Studio Cam goes live

    Broadcast on NOUZIK

    Six-hour open-studio livestream — three producers working in real time. Viewers vote on the final mix.

  4. 14:00

    Introduction to AI in Music

    Caudan Arts Centre · Talk Hall

    Stems separation, voice modelling, MIDI co-writing, and the line between tool and author. Ethics included.

  5. 15:30

    Sample Ethics with Traditional Material

    Caudan Arts Centre · Talk Hall

    Working with field recordings, family archives and the Heritage Archive — credit, consent, royalties.

  6. 17:00

    NOUZIK Platform Demo

    NOUZIK Studio · Demo room

    From upload to release in 30 minutes. Royalty splits, playlists, analytics, livestream tooling — walkthrough by the team.

  7. 19:00

    Producers' Listening Session

    Caudan Arts Centre · Main Hall

    Eight producers play one track each on the big system. Honest feedback from the room. No phones.

  8. 21:00

    AI × Creole Sound Showcase

    Caudan Arts Centre · Main Hall

    BerTz headlines a hybrid set: live ravanne, vocals from Melanie Amboule, AI-assisted production on stage.

AI as a co-writer

Tools that suggest chords, separate stems and clone style — used in public, with the producer in charge. Today we show what works and what we refuse to ship.

From session to release

Every track polished today can be uploaded to NOUZIK before midnight. Distribution, splits and analytics handled in the same room.

Open studio, all day

The Studio Cam livestream means anyone, anywhere can watch producers work — chat is open, the mix is interactive, the music is real.

On stage today

Featured artists.

  • BerTz
    DJ · Producer · Lecturer · In-house resident
  • Melanie Amboule
    Live vocals & instrumentation
  • Rémi Beaugendre
    Percussion · session
  • Damien Pontil
    Guitar · session
Voices of the day

Speakers & mentors.

  • BerTz
    Producer & Lecturer

    Leads the advanced roundtable and headlines the AI × Creole Sound showcase with an exclusive live set.

  • NOUZIK Platform Team
    Distribution & rights

    Walking through the full release flow — upload, splits, livestream, royalties — with real artist accounts on screen.

Quick tutorials

Take this home. Try it tonight.

Compact, step-by-step walk-throughs distilled from today's sessions.

Build your first séga-tipik flavoured loop

  1. 01Open your DAW at 100 bpm, 6/8 time signature.
  2. 02Drop a ravanne sample on beat 1 of bar 1 — that's your anchor 'dab'.
  3. 03Layer a maravanne shaker on every off-pulse — washing, continuous, sidechained -3 dB to the ravanne.
  4. 04Add a triangle hit on the last pulse of bar 2 — that's your accent.
  5. 05Loop 8 bars. Add a vocal chop or a guitar skank on bars 5–8 only. Bounce. You have your first sketch.

Use AI as a co-writer (without losing your voice)

  1. 01Hum your idea into a voice memo first. Always start with your sound.
  2. 02Use a tool like Suno / Udio only to expand the idea — never to start it.
  3. 03Generate variations, then keep only the elements you'd have written anyway.
  4. 04Re-record every kept element yourself, with real instruments or vocals.
  5. 05Document the prompts and outputs you used. Future you needs receipts.

Release a track on NOUZIK in 30 minutes

  1. 01Have a stereo master (WAV, -14 LUFS integrated), cover art (3000×3000 px), and your split sheet ready.
  2. 02Open NOUZIK Studio → New Release → upload audio + art.
  3. 03Add metadata: title, ISRC (auto-assigned if you don't have one), genre, language.
  4. 04Enter the splits exactly as they appear on the signed split sheet. Each collaborator confirms by email.
  5. 05Pick a release date (minimum +7 days for editorial pitching), submit. You're live.
Resources

Keep going after the room empties.

Glossary

The words you heard today.

DAW (Digital Audio Workstation)
Software you record, arrange, mix and produce in. Free options: GarageBand (Mac/iOS), Cakewalk by BandLab (Windows). Paid: Ableton Live, FL Studio, Logic Pro.
Audio interface
Hardware that converts analogue sound (voice, guitar) into digital audio your computer can record — and back to analogue for monitoring. Usually the single biggest quality upgrade from a laptop's built-in jack.
Studio monitors
Speakers built for flat, honest reproduction. Consumer speakers are tuned to sound 'exciting' — they hide mix problems.
MIDI controller
A keyboard or pad controller that sends performance data (not audio) into the DAW. Lets you program drums and melodies without recording real sound.
Signal flow
The path sound takes: source → microphone → preamp / interface → DAW → mix bus → master → file. Knowing the chain is how you debug any problem.
Sample rate
How many times per second audio is measured in digital form (44.1 kHz or 48 kHz). Streaming delivery standard is 44.1 kHz / 16-bit.
Bit depth
How much detail is captured in each digital audio measurement (16-bit or 24-bit). Higher = more headroom while mixing.
EQ (Equalisation)
Boosting or cutting specific frequency ranges to shape tone or carve space for each element in a mix.
Compressor
Evens out volume. Four controls: threshold (level it starts working), ratio (how strongly it reduces), attack (how fast it reacts), release (how fast it stops).
Sidechain
Routing one signal to control the volume of another — most commonly, a kick drum ducks the bass for clarity on every beat.
Stem
An isolated track of one element (vocal, drums, bass…). Used for mixing, remixing, or extracted with AI separation.
LUFS
Loudness Units Full Scale — the perceptual loudness measurement streaming platforms use. Streaming targets ≈ -14 LUFS integrated.
Mastering
The final polish stage — loudness, EQ balance and format export for distribution across phones, clubs and headphones alike.
BPM (Beats Per Minute)
A precise numerical measure of tempo. Central to both production (project tempo) and DJing (beatmatching).
Beatmatching
The foundational DJ skill: adjusting the tempo and timing of one track to align with another for a seamless transition.
Phrasing
The structural grouping of music into recognisable blocks (e.g. 8 or 16-bar sections). DJs mix on phrase boundaries, not random beats.
Plugin / VST
Software add-ons used within a DAW to process or generate sound (an EQ plugin, a virtual instrument).
Sample
A pre-recorded piece of audio reused inside a new track. Always requires clearance if it comes from someone else's recording.
MIDI
Musical Instrument Digital Interface — note data (not audio) you can edit and re-assign to any instrument.
Voice modelling / cloning
AI that learns a voice's characteristics and generates new audio in that voice. Without the performer's explicit consent, this is a hard ethical boundary — not a grey area.
Stems separation (AI)
AI tool that splits a finished mix back into vocals / drums / bass / other. Useful for remixes and educational analysis.
Royalty split
Agreed division of future income from a track between all contributors. Should be locked before release — covered in detail on Day 4.
"Technology is not the destination. It is a new instrument. The song is still ours to write."
BerTz · Production Day Keynote