MusicLM

MusicLM Overview

MusicLM is a high-fidelity music generation model developed by Google Research. It generates music from text descriptions, casting the process of conditional music generation as a hierarchical sequence-to-sequence modeling task. The model can generate music at 24 kHz that remains consistent over several minutes. It outperforms previous systems in terms of audio quality and adherence to the text description. MusicLM can also be conditioned on both text and a melody, transforming whistled and hummed melodies according to the style described in a text caption.

MusicLM Highlights

• MusicLM generates high-fidelity music from text descriptions, maintaining consistency over several minutes.
• The model outperforms previous systems in terms of audio quality and adherence to the text description.
• MusicLM can transform whistled and hummed melodies according to the style described in a text caption.