MusicLM: Generating Music From Text

US Google model makes high-fidelity music from descriptions 29/01/23

A.I. has taken great strides in recent years, you only have to look at the internet for 10 seconds to see someone using ChatGPT to generate a picture of a cat with an imaginary vintage synth or something equally amusing! Here is another novel method, straight from the Google Labs - using text to generate music. The results are, as you can imagine, slightly spooky! More details here:

Abstract We introduce MusicLM, a model generating high-fidelity music from text descriptions such as "a calming violin melody backed by a distorted guitar riff". MusicLM casts the process of conditional music generation as a hierarchical sequence-to-sequence modeling task, and it generates music at 24 kHz that remains consistent over several minutes. Our experiments show that MusicLM outperforms previous systems both in audio quality and adherence to the text description. Moreover, we demonstrate that MusicLM can be conditioned on both text and a melody in that it can transform whistled and hummed melodies according to the style described in a text caption. To support future research, we publicly release MusicCaps, a dataset composed of 5.5k music-text pairs, with rich text descriptions provided by human experts.

Find out more on the Google research page: https://google-research.github.io/seanet/musiclm/examples/

Here's another interesting example, with sound effects generated by A.I. from silent video. Bear in mind this is 6 years old!

About the author [midierror]: midierror makes nifty Max For Live devices, innovative music hardware, award winning sample packs and hosts a podcast speaking to people in the music world.