MusicLM: Turn Ideas Into Music

MusicLM is a model that can generate music from text descriptions, such as “a calming guitar melody in 6/8 time signature riff”. It was developed by Google Research and can produce high-fidelity music at 24 kHz that remains consistent over several minutes. MusicLM uses a hierarchical sequence-to-sequence approach to model the structure and style of music. It can also be conditioned on both text and a melody, such as a whistled or hummed tune, and transform it according to the text caption. MusicLM is an experimental tool that aims to empower the creative process of musicians and music enthusiasts.

✅ Audio Generation From Rich Captions

✅ Long Generation

✅ Story Mode

✅ Text and Melody Conditioning

✅ Painting Caption Conditioning

✅ 10s Audio Generation From Text

✅ Generation Diversity