All articles tagged with #ai music generation
Originally Published 2 months ago — by Engadget
OpenAI is developing an AI tool that can generate music from text and audio prompts, collaborating with Juilliard students to create training data, aiming to produce features like guitar accompaniment and video music addition.
Originally Published 5 months ago — by TechCrunch
ElevenLabs has launched an AI music generator that is claimed to be cleared for commercial use, expanding beyond its traditional text-to-speech tools. The company has shared samples of AI-generated music and announced partnerships with major music publishers to use their material for training, amid ongoing legal concerns about copyright infringement in AI music development.
Originally Published 5 months ago — by Music Business Worldwide
ElevenLabs has launched Eleven Music, an AI platform for generating studio-grade music from natural language prompts, with licensing deals with Kobalt and Merlin to protect rights holders and ensure fair compensation, entering the competitive AI music space with a focus on responsible licensing and industry collaboration.
Originally Published 2 years ago — by WIRED
Google DeepMind has unveiled its most advanced AI music generation model called Lyria, along with two music-making experiments. One experiment allows users to transform hummed melodies or keyboard solos into guitar riffs or choir arrangements, while the other, called Dream Track, enables users to create 30-second YouTube Shorts using AI-generated voices and musical styles of popular artists like T-Pain, Sia, Demi Lovato, and Troye Sivan. However, concerns arise regarding the ease with which AI can create music, potentially undermining the creative process. Additionally, copyright issues and the impact on artists' work and revenue are being raised as generative AI becomes more prevalent.
Originally Published 2 years ago — by Engadget
Meta's Audiocraft research team has released MusicGen, an open-source deep learning language model that can generate new music based on text prompts and even be aligned to an existing song. MusicGen is faster than existing methods and can be run locally, available in four model sizes. It is open source and can even be used to generate commercial music.
Originally Published 2 years ago — by Music Business Worldwide
Google's experimental AI tool, MusicLM, which generates high-fidelity music from text prompts, is now available for public use. The tool was trained on five million audio clips and can create two versions of a requested song for users to vote on. However, Google has highlighted the risks associated with generative AI in music, including cultural appropriation and copyright infringement. The release of MusicLM comes amid rising concerns around the use of generative AI in music and the potential for streaming fraud.