• btc = $57 966.00 607.59 (1.06 %)

  • eth = $3 129.35 27.71 (0.89 %)

  • ton = $7.36 -0.00 (-0.04 %)

  • btc = $57 966.00 607.59 (1.06 %)

  • eth = $3 129.35 27.71 (0.89 %)

  • ton = $7.36 -0.00 (-0.04 %)

30 Jan, 2023
1 min time to read

MusicLM creates compelling pieces using captions, scripts, or even hummed melodies.

Google has developed a new AI system called MusicLM that can turn text into music. The AI system can produce minutes-long musical pieces from text prompts and can even transform a whistle or hummed melody into other instruments. The company has released a research paper detailing the technology behind MusicLM and has uploaded a number of samples produced using the model to showcase its capabilities.

MusicLM uses a deep learning approach to generate music from text, which allows it to produce a wide range of musical styles, from classical to electronic. The system can also generate music in response to specific instructions, such as the genre, vibe, or instruments that are to be used. Additionally, MusicLM can simulate human vocals and has demonstrated its ability to transcribe a whistle or hummed melody into different musical instruments.

Google is being cautious with MusicLM, citing the risks of "potential misappropriation of creative content" and potential cultural appropriation or misrepresentation. The company has no plans to release the model to the public at this time. However, it is publicly releasing a dataset of around 5,500 music-text pairs, which could help researchers and AI developers train and evaluate other musical AI systems.

Overall, MusicLM is a remarkable achievement in the field of AI-generated music and offers a glimpse into what the future of music production could look like. With further development, it's possible that AI systems like MusicLM could play a significant role in the creation of new music, freeing up musicians to focus on other aspects of the creative process.