How to make music with AI (step by step)

To make music with AI, open wawie's music generator and describe the track in one sentence: genre, mood, instruments, tempo. Pick a genre and a mood, keep the instrumental option on for background music, and generate. Compare the takes, then download your favourite as an MP3 or drop it into a video edit. The music is original and royalty-free.

Create a free account 5,000 free credits. No card needed.

Updated September 27, 2026

An AI music generator composes an original track from a text description. You do not need an instrument, music theory or a stock library subscription. In wawie you describe what you want, set a genre and a mood, and get finished takes you can compare, download or place under a video. Because each track is generated for you rather than taken from an existing catalogue, it is royalty-free to use within the model providers' terms.

Time needed: 10 min

What you need

A free wawie account
A one-sentence description of the track
A genre and mood in mind
The video or podcast it is for (optional)

How do you make music with AI, step by step?

Create a free wawie account: Sign up in your browser, with nothing to install and no card needed. The free plan includes 5,000 one-time credits and limited music generation, so you can try the generator first. Paid plans, from Wawie Start at EUR 4.99 a month, add full music generation and commercial use.
Open the music generator: From the side menu of the dashboard, open the music tool. It has a simple mode that works from a description, and an advanced mode that adds fields for lyrics, styles and a song title. For background music, the simple mode is all you need.
Describe the track in one sentence: Write what you want to hear: genre, mood, instruments, tempo and where it will be used. For example, a warm lo-fi beat for a focus playlist, or a mysterious orchestral bed for a documentary. If you are stuck, start from one of the inspiration ideas in the settings.
Pick a genre and a mood: Choose a genre, from pop and lo-fi to cinematic, jazz or afrobeat, and a mood such as uplifting, calm, dark or epic. Both are added to your description, so the model receives a precise brief instead of a vague one. Leave them open if you want more variety.
Keep the instrumental option on: For background music under a voice, leave the instrumental option switched on so the track has no vocals. You can also list styles to avoid, such as heavy metal or distortion, to steer the result away from sounds that would clash with your video.
Generate and compare the takes: Launch the generation. The cost is shown before you start, and a full track takes from tens of seconds to a few minutes. Each run can return more than one take, so listen to them all in the player and keep the one that fits best.
Refine the description and regenerate: If a take is close but not quite right, change one thing at a time: the tempo, an instrument or the mood. Small, precise edits to the description move the result more reliably than rewriting the whole prompt, and each new run gives you fresh takes.
Download it or score your video: Download the track as an MP3, or find it in your media library and drag it into a studio project. In the studio, automatic ducking lowers the music under the voice, and the finished edit exports as an MP4 ready for YouTube, TikTok or Instagram.

How do you get better AI music?

Name the use in your description. A podcast intro, a trailer build and a vlog opener each call for a different energy and structure, so say which one the track is for.
Be specific about instruments and tempo. Asking for slow piano and soft strings gives the model far more to work with than a vague request for calm music.
Use styles to avoid to rule out what you do not want, such as distortion, brass or heavy drums, instead of hoping the model leaves them out.
For full control of the arrangement, open the Music Lab in the studio. It generates separate stems, such as drums, bass and melody, that you mix yourself on the timeline.

Make your first track with wawie's AI music generator

Frequently asked questions

Is AI-generated music really royalty-free?

Tracks you generate in wawie are composed for you and come from no existing catalogue, so there is no per-track license to buy. On a paid plan, from Wawie Start, you can use them in your projects, including commercial ones, within the underlying model providers' terms. You remain responsible for anything you add yourself, such as lyrics or samples.

How long does it take to generate a track?

Usually from tens of seconds to a few minutes, depending on the length of the track and how busy the model provider is. Progress is shown while you wait. In practice, writing a precise description takes longer than the generation itself, and it is also the part that most improves the result, so it is worth a minute of thought.

Can I use AI music in YouTube videos?

Yes. Music generated in wawie is original and royalty-free, which suits background music for YouTube, TikTok, Instagram and podcasts, and commercial use is included from Wawie Start. Keep the downloaded file and a note of when you created it, so you can show where the track came from if a platform ever asks about it.

What is the difference between AI music and an AI song?

Both come from the same wawie tool. With the instrumental option on, you get music without vocals, suited to videos, podcasts and presentations. Switch it off and the track is sung: describe a song and its words are written for you, or use the advanced mode to supply your own lyrics. Song creation is included from Wawie Start.

How much does it cost to make music with AI?

Music is billed in credits per generation, and the cost is shown before you launch. The free plan includes 5,000 one-time credits with no card and limited music generation. Paid plans start with Wawie Start at EUR 4.99 a month for 180,000 credits a month, and each plan also covers voices, dubbing, sound effects and the video editor.

Keep exploring

Create a free account 5,000 free credits. No card needed.