Google DeepMind describes a music generation model and experiments intended to explore new ways to create music.
MusicLM
MusicLM explores ways to generate music based on written descriptions.
About MusicLM
MusicLM is a Google Research model that generates music from text descriptions. The page presents example outputs, including generation from rich captions, long-form passages that remain consistent over several minutes, and 10-second samples organized by instrument, genre, place, epoch and player experience level.
It also demonstrates Story Mode, which continues audio across a sequence of text prompts, conditioning on both text and a whistled or hummed melody, and conditioning on painting descriptions. The page links to the research paper and to MusicCaps, a dataset of 5.5k music-text pairs with descriptions written by human experts.
Key features
- Generate music at 24 kHz from text descriptions
- Maintain consistency over several minutes in long generation
- Continue audio across sequence of text prompts in Story Mode
- Transform whistled or hummed melodies according to text style
- Generate music conditioned on painting title, image and description
- Release MusicCaps dataset with 5.5k music-text pairs
Best for: Researchers and others reviewing examples of text-described music generation and the associated MusicCaps dataset.
Good for
- Provides audio examples for many prompt types including instruments and genres
- Documents long-form, story-mode and melody-plus-text generation approaches
- Releases MusicCaps dataset with expert-written descriptions for future research
MusicLM at a glance
- Category
- AI & Machine Learning › AI Music Generation
- Type
- Website
- Pricing
- Not confirmed. Check the site
- Platforms
- Web
- Open source
- Not stated
- Last checked
- Oct 5, 2026 · Online when checked
- Listed
- Oct 5, 2026
MusicLM alternatives
Tools and sites you could use instead, ranked by how closely they match.
Generate a soundtrack matching an uploaded image, with controls for denoising steps and eta.
Ecrett Music lets users create royalty-free music tracks through its online service.
Examples for ICASSP2024 paper “StemGen: A music generation model that listens”.
Google Flow Music supports creating, remixing, and sharing songs with generative AI.
Yue AI generates full-length songs from lyrics, including vocals, instrumentals, and arrangements.
Popular in AI Music Generation
Well-known listings from the same topic.
Suno AI Musicsunoai-music.comMusicCreator AImusiccreator.aiMubertmubert.comAudioXaudiox.appAIVAaiva.aiAisongmakeraisongmaker.io
MusicLM FAQ
What is MusicCaps?
A publicly released dataset of 5.5k music-text pairs with rich descriptions provided by human experts.
Can MusicLM combine text with a melody?
Yes, it can transform whistled and hummed melodies according to the style described in a text caption.
How does Story Mode work?
Audio is generated from a sequence of text prompts that influence how the model continues semantic tokens from the previous caption.
Is something wrong or out of date?
Tell us and an editor will check it.
Own this website?
Claim this listing
Claim the listing and suggest updates yourself. An editor checks every claim first.