Google researchers have created a new AI tool that can create melodies from text.
Its similar to how DALL-E can generate images from written prompts and ChatGPT can turn a text command into story.
The AI program called MusicLM can turn text input into seconds, and even minutes-long music.
Google has shared some samples that it produced using MusicLM
The samples are called MusicCaps and are basically a dataset composed of 5.5k mu ..
The company said, 'MusicLM generates music at 24 kHz that remains consistent over several minutes.'
The examples shared by the company included 30 second clips and 5 minute clips that sound like actual songs.
The researchers said, 'The audio is generated by providing a sequence of text prompts.'
There is also a 'story mode' demo where the model is basically given multiple text inputs with time duration for each type of music that needs to be created.
The company claims that MusicLM outperforms previous systems.