Generate Music

Compose original soundtracks in seconds. WaveSpeed utilizes advanced MusicGen and Suno AI models to turn text descriptions into high-fidelity audio. From lofi beats to cinematic orchestral scores, generate royalty-free music tailored to your exact duration and mood requirements.

Generate Music Now View API Docs

Music Generation Capabilities

Explore the different ways to generate music using our unified API.

1. Full Track Generation

Powered by Suno AI / MusicGen-Large. Generate complete songs with structured verses, choruses, and bridges. Capable of understanding complex musical theory concepts like "120 BPM," "Key of C Minor," or "Jazz Fusion style." Best for background music for videos, complete songs for streaming. Pair generated music with Audio for Video workflows for complete production pipelines.

2. Loop & Stem Creation

Powered by AudioCraft / Riffusion. Create seamless, infinite loops or isolated instrument stems (e.g., just the drum beat or just the bassline). Perfect for game developers needing adaptive audio layers. Best for game assets, beat-making, and sampling. Available on WaveSpeed.

3. Jingle & Short Audio

Powered by MusicGen-Melody. Generate short, memorable audio clips based on a melody you hum or a text description. High adherence to timing constraints makes it ideal for intros and outros. Best for podcast intros, logo reveals, and commercial transitions. Combine with video generation models for end-to-end content creation.

Text-to-Music Examples

See how text prompts translate into audio reality.

Cinematic Background Score

Prompt: "Epic orchestral soundtrack, building tension, 90 BPM, cinematic strings with timpani crescendo." Generates a 30-second score perfect for trailers and documentary intros. Pair with Audio for Video for seamless integration.

Lo-Fi Study Beat

Prompt: "Chill lo-fi hip hop beat, vinyl crackle, mellow piano chords, 75 BPM, infinite loop." Creates a seamless loop ideal for livestream backgrounds, study playlists, and ambient content.

Podcast Intro Jingle

Prompt: "Upbeat, energetic podcast intro, acoustic guitar riff, 5 seconds, catchy and memorable." Produces a short, punchy jingle tailored for brand recognition across episodes.

Q & A

How does AI Generate Music?

AI music models (like Transformer-based architectures) analyze patterns in vast datasets of music to understand harmony, rhythm, and structure. When you provide a text prompt, the AI predicts and generates audio waveforms that match your description.

Is the generated music royalty-free?

Yes. Music generated on WaveSpeed is unique and royalty-free. You can use it in monetized YouTube videos, podcasts, and commercial projects without fear of copyright strikes.

Can I generate vocals?

Some advanced models like Suno AI support vocal generation, including lyrics and singing in various styles. Other models like MusicGen focus primarily on instrumental tracks. Check the specific model capabilities in our documentation.

What is the maximum duration?

This depends on the model. Standard generation usually ranges from 30 seconds to 2 minutes per clip. However, our "Continue" feature allows you to extend a track indefinitely by using the end of the previous clip as context for the next segment.

Can I reference an existing melody?

Yes. We support "Melody Conditioning." You can upload a short audio file (like a hummed tune or a whistle) and the AI will generate a full arrangement that follows your melody line.