Song Creator Pro is an AI music generator that runs entirely on your Windows PC. You type a description, add lyrics, hit generate, and get a full song back — vocals, instruments, and all. No musical experience required.

This guide walks you through creating your first song in five steps. By the end, you'll have a finished AI-generated track and know how to make more.

Step 1: Write Your Description

The Description field on the left side of the interface is where you tell the model what kind of music to make. This is the single most important input — the more detail you include, the better your results.

Cover as many of these as you can:

Genre — pop, rock, jazz, lo-fi, EDM, country
Instruments — acoustic guitar, piano, synth pads, 808 drums
Mood — melancholic, uplifting, energetic, dreamy
Vocal style — female vocal, male vocal, breathy, powerful, raspy
Production feel — polished, raw, lo-fi, live recording

A vague description like "a happy song" gives the model too much freedom and you'll get unpredictable results. Be specific.

Example description you can copy:

upbeat indie pop, acoustic guitar, bright piano, male vocal, energetic and carefree, polished production, catchy melody, summer vibes

Turn on AI Enhance. It's the toggle next to the Description label. It reformats your description into the structure the AI model understands best. You still write naturally — AI Enhance handles the formatting.

Step 2: Add Your Lyrics

The Lyrics field in the center is where you write the words for your song. The key to good results is breaking your lyrics into sections using tags like [Verse], [Chorus], and [Bridge].

These tags tell the model how to structure the song's energy — verses are calmer, choruses hit harder, bridges shift the feel.

Example lyrics you can copy:

[Verse 1]
Woke up to the sound of rain
Sun broke through the windowpane
Grabbed my keys and hit the road
Shook off everything I owed

[Chorus]
This is the start of something good
Doing all the things I said I would
No more waiting, no more standing still
If I don't move now, I never will

[Verse 2]
Radio playing our old song
Windows down and singing along
Every mile puts it all behind
Leaving all the noise inside my mind

[Chorus]
This is the start of something good
Doing all the things I said I would
No more waiting, no more standing still
If I don't move now, I never will

[Outro]
If I don't move now, I never will

Tips for lyrics:

Keep it to roughly 2-3 words per second of audio. For a 2-minute song, aim for 90-140 total words. Too many words and the vocals will sound rushed.
Use [Instrumental] between sections to add breathing room.
For a purely instrumental track, just leave the lyrics field empty.

Step 3: Set Music Details (Optional)

Below the lyrics field, you'll see Music Details: BPM, Duration, Key, Time Signature, and Language.

For your first song, leave these on their defaults. The model is smart enough to pick appropriate values based on your description and lyrics. If you do set them manually, make sure they match your description — setting BPM to 150 while asking for a "slow ballad" will send mixed signals and hurt the output.

When you do want manual control:

Setting	What it does	When to set it
BPM	Tempo (beats per minute)	When you need a precise tempo for syncing to video or other tracks
Duration	Track length in seconds (10s–600s)	When you need a specific length (default is fine otherwise)
Key	Musical key (e.g. C major, A minor)	When combining with other tracks that need to be in the same key
Time Sig	Time signature (e.g. 4/4, 3/4)	For waltz (3/4) or unusual meters
Language	Vocal language	When singing in a language other than English

Step 4: Check Advanced Settings

Click Advanced Settings in the bottom-left to open the settings panel. Here are the ones that matter most:

Steps — How many passes the model makes to refine your audio. Higher = better quality but slower. 32 is a good default for the SFT model. For the turbo model, leave it at 8.
Guidance — How strictly the model follows your description. 5.0-7.0 is the sweet spot. Below 1.5 it ignores your prompt. Above 12 you get audio artifacts.
Batch Size — How many versions to generate at once. Set this to 2 or higher. AI music has inherent randomness, so generating multiple versions and picking the best one is the most effective workflow.
Seed — Set to -1 for random (default). Use a specific number to reproduce a previous result.

Leave the other settings (Shift, LM CFG Scale) at their defaults for now.

Step 5: Generate and Pick Your Best Result

Hit the Generate button in the top-right corner. You'll see a progress bar as the model works through the inference steps.

Once generation is complete, your results appear in the Results panel on the right. If you set batch size to 2, you'll get two variations to compare. Listen to each one and use the navigation arrows to flip between them.

For each result, you have three options:

Load — Loads that result's settings back into the editor so you can tweak and regenerate
Remix — Uses that audio as a starting point for a new variation
Save — Exports the track to your computer

The iteration workflow: Don't expect perfection on the first try. Generate 2-4 versions, pick the best one, then adjust your description or lyrics and generate again. This is how you get great results consistently.

Your First Song: Try This

Here's a complete example you can paste directly into Song Creator Pro to generate your first track right now:

Description:

upbeat indie pop, acoustic guitar, bright piano, male vocal, energetic and carefree, polished production, catchy melody, summer vibes

Lyrics:

[Verse 1]
Woke up to the sound of rain
Sun broke through the windowpane
Grabbed my keys and hit the road
Shook off everything I owed

[Chorus]
This is the start of something good
Doing all the things I said I would
No more waiting, no more standing still
If I don't move now, I never will

[Verse 2]
Radio playing our old song
Windows down and singing along
Every mile puts it all behind
Leaving all the noise inside my mind

[Chorus]
This is the start of something good
Doing all the things I said I would
No more waiting, no more standing still
If I don't move now, I never will

[Outro]
If I don't move now, I never will

Settings: Leave everything on defaults. Turn on AI Enhance. Set batch size to 2. Hit generate.

What's Next

Once you've made your first song, check out our AI Music Prompting Guide for a deeper dive into description writing, advanced lyric formatting, and how to get the most out of every setting.

Ready to start? Get Song Creator Pro on the Microsoft Store for $49.99 — one-time purchase, unlimited AI music generation, runs entirely on your PC. Also available on itch.io for $44.99.

How to Make Your First AI Song with Song Creator Pro