Module 08: AI Tools & Workflow

The Problem with Bad Prompts

"Make me a song" - that's like going to a restaurant and saying "Bring me food." You'll get what the chef likes, not what you want.

AI is a mirror: It shows you how precisely you think. A bad prompt leads to generic results. A good prompt leads to exactly what you imagined - or even better.

❌ Bad Prompt

A beautiful song about love

Vague, not specific, no direction

✅ Good Prompt

[Synthwave, 110 BPM] Instrumental with pulsating analog bass, glittering arpeggios, nostalgic 80s feeling. Structure: Intro → Build-up → euphoric Drop → fading Outro. Mood: melancholic euphoria, like a night drive through neon lights.

Genre Tempo Instruments Mood Structure

The PANTAM Prompt Formula

After hundreds of generated songs, we have developed a proven structure:

The 5-Element Formula

[GENRE, TEMPO] INSTRUMENTS with MOOD/ATMOSPHERE. STRUCTURE.

The AI-Assisted Music Workflow

Concept & Vision

Detailed Instructions

Suno/ElevenLabs

Audacity/DAW

Export & Release

1. Genre (The Foundation)

Choose a specific sub-genre instead of the umbrella term:

Instead of "Electronic" → "Deep House", "Techno", "Synthwave", "Drum & Bass"
Instead of "Rock" → "Indie Rock", "Post-Punk", "Stoner Rock", "Math Rock"
Instead of "Hip Hop" → "Boom Bap", "Trap", "Lo-Fi Hip Hop", "Grime"

2. Tempo (The Pulse)

Give specific BPM (Beats Per Minute):

60-80 BPM: Ballads, Triphop, Ambient
90-100 BPM: Hip Hop, Pop Ballads
110-130 BPM: House, Pop, Disco
140-175 BPM: Drum & Bass, Hardstyle, Punk

3. Instruments (The Colors)

Describe not just which, but how they should sound:

"distorted electric guitar" vs. "clean acoustic guitar"
"punchy analog synthesizer" vs. "glittering digital pads"
"crispy trap drums" vs. "warm acoustic drums"

4. Mood/Atmosphere (The Emotion)

Use vivid language:

"melancholic euphoria like a sunset at the beach"
"dark tension like in a film noir"
"playful lightness like a spring morning"

5. Structure (The Dramaturgy)

Define the flow:

"Intro → Verse → Chorus with Breakdown → Bridge → Final Chorus"
"Steady Build-up over 2 minutes, then Drop"
"ABAB structure, each B more intense than the previous"

Advanced Prompting: Metatags

Some AI tools understand special tags. Here are the most important ones for Suno:

Tag	Description	Example
`[Instrumental]`	No vocals, only music	[Synthwave, 120 BPM, Instrumental]
`[Male Vocal]` / `[Female Vocal]`	Singer gender	[Pop, Female Vocal, emotional]
`[Whisper]` / `[Belting]`	Vocal style	[Alternative, Whisper, intimate]
`[Reverb-heavy]`	Lots of reverb/echo	[Ambient, Reverb-heavy, spacious]
`[Lo-fi]`	Vintage, noisy sound	[Hip Hop, Lo-fi, dusty]
`[Live Recording]`	Sounds like live recording	[Jazz, Live Recording, raw]

🎵 Exercise: Prompt Upgrade

Take this weak prompt and apply the 5-Element Formula:

Original: "A sad song"

Your Version: [Write your optimized prompt here]

Example solution: [Indie Folk, 75 BPM] Acoustic guitar with fingerpicking, gentle cello, dreamy piano. Mood: autumn melancholy like a walk through rainy streets. Structure: Verse with increasing instrumentation → emotional Chorus → fading Outro with solo cello.

The AI Tool Landscape 2026

AI is evolving rapidly. Here are the most important tools for audio production:

☀️

Suno AI

Complete Song Generation

Generates complete songs with vocals, instruments and arrangement from text prompts.

50 credits/day (Free)
Stems download available
Up to 4 minutes length
Various styles

🎶

Udio

Alternative to Suno

Similar to Suno, but with focus on higher audio quality and better vocal sound.

600 credits/month (Free)
Longer songs possible
Stronger at complex genres
Inpaint feature for edits

🗣️

ElevenLabs

Voice Cloning & TTS

Create realistic voices, clone your own voice, generate voice-overs.

10,000 characters/month (Free)
29+ languages
Voice cloning with 3 minutes audio
Emotion control

🎼

AIVA

AI Composition

Focus on classical orchestration and film music. MIDI export for DAWs.

3 downloads/month (Free)
MIDI export
Sheet music view
Style references

🎸

Moises

Stem Splitting

Separates existing songs into stems (Vocals, Drums, Bass, etc.). Perfect for remixing.

5 uploads/month (Free)
4-stem splitting
Metronome detection
Pitch/Chord detection

🎧

LALAL.AI

Advanced Stem Splitting

High-quality stem splitting with up to 10 stems (including guitars, synths).

10 minutes (Free)
10-stem splitting
Voice Cleaner
Batch processing

Professional AI Workflows

Workflow 1: The Rapid Prototyper

For songwriters who want to quickly test ideas:

Step 1: Idea

Write lyrics (1 Verse + Chorus)
Define genre & mood
Note 3 reference artists

Step 2: Generate

Create 5 variants in Suno
Choose best version
Note what works

Step 3: Decide

Is the idea good? → Continue in DAW
Or new direction? → New prompt

Workflow 2: The Sound Designer

For producers looking for unique sounds:

Step 1: Collect

Generate 10 instrumentals
In different genres
With unusual combinations

Step 2: Extract

Download stems
Import into DAW
Sample interesting loops

Step 3: Combine

Drums from Song A
Bass from Song B
FX from Song C

Workflow 3: The Remix Artist

For creative remixes and mashups:

Step 1: Sources

Search for acapellas (YouTube, etc.)
Or: ElevenLabs for vocals
Note BPM and Key

Step 2: Instrumental

Generate matching instrumental
Match BPM and Key
Download stems

Step 3: Mix

Import into DAW
Time-stretch if needed
Mix together

Ethics & Copyright

🎵 The Golden Rule of AI Usage

AI is a tool, not a replacement for creativity. You remain the artist - the AI is your brush, not the painter.

Be transparent: Say when you use AI
Transform: Use AI as a starting point, not an endpoint
Respect: Don't use voices of real people without permission
Learn: Understand what the AI does, don't blindly trust
Create: Always add your own touch

What is allowed?

Your own prompts: The generated music belongs to you (with Suno/Udio Free tier: non-commercial; Pro: commercial allowed)
Remixing stems: As long as you transform and don't 1:1 copy
Voice cloning: Only your own voice or with explicit permission

What is problematic?

Deepfakes: Cloning other voices without permission
1:1 Uploads: Passing AI-generated songs as your own without modification
Trademark Violations: Prompts like "in the style of [famous artist]" for commercial releases

Tip: The music industry is evolving. Current guidelines: Suno and Udio allow commercial use with paid plans. When in doubt: Transparency is the best protection.

The Future of Music Production

AI is changing the music world - but not as many fear. It democratizes production: Everyone can now make professional-sounding music, regardless of budget or studio access.

This still distinguishes pros from amateurs:

Taste: What sounds good? What fits the context?
Decisions: Which of the 10 AI variants is the right one?
Refinement: How do you transform an AI seed into a finished track?
Storytelling: What do you want to say with the music?

🎵 Final Project: Your AI Portfolio

Show what you've learned:

Create 3 songs in different genres (e.g., Lo-Fi, EDM, Indie)
Use the 5-Element Formula for each
Load the stems into your DAW
Change at least 3 elements per song (arrangement, mix, own recordings)
Export all 3 as WAV
Write a short "production note" for each song (what did you do?)

Upload: Upload your favorite tracks in the Audio Player!

What's Next?

In Module 09 you will learn mastering - the final polish that makes your tracks release-ready.

Reminder: AI is an accelerator, not a magic wand. The best results come from people who know what they want - and use AI as a tool to get there faster.