Beatoven.ai Unveils Maestro AI Music Generation Model, Announces Revenue-Sharing for Artists

Beatoven.ai, a Bengaluru-based artificial intelligence (AI) startup focused on music generation, unveiled its first in-house AI Model on Wednesday. Dubbed Maestro, it is a music model that can generate instrumental tracks across various genres based on text prompts. The company claims that the large language model (LLM) was only trained on fully licensed datasets, and the rights holders of the music are also eligible for revenue-sharing. The Maestro AI model will be available to the paid subscribers of the platform.

Beatoven.ai Introduces AI Music Generation Model

In a press release, the Indian AI startup announced and detailed its latest foundation model. Its unique selling proposition (USP) remains being trained on only licensed content, which means the output generated from the model can be used commercially without the fear of it plagiarising another artist or facing a copyright infringement lawsuit.

Currently, the model can generate instrumental tracks from text prompts, and the company promises that sound effects and vocals will be added at a later date. Interestingly, this is the first such AI model by the company that can generate music from scratch. Its earlier models were essentially a composition system, where the AI would combine different music samples from partnering artists and create new soundtracks.

In a conversation with Gadgets 360, Beatoven.ai CEO Mansoor Rahimat Khan said, “Maestro is trained on 30X the dataset of the previous models, and the AI model can generate instrumental tracks with higher quality and across diverse genres.”

Coming to the revenue-sharing aspect, Beatoven.ai collaborated with Musical AI for dataset licensing, a rights management platform that allows AI companies to attribute generative outputs to specific tracks and pay music rightsholders. It has also partnered with other data providers, including Rightsify, Soundtrack Loops, Symphonic Music, Bobby Cole, Vadi Sound, and Pro Sound Effects.

According to Khan, revenue-sharing works in a straightforward manner. Whenever the generated output contains elements that it learned from an artist's sample, the artist will be paid a part of the revenue generated by the company. Musical AI handles the attribution for transparency's sake. Khan told Gadgets 360 that artists will receive 30 percent of the total revenue generated by them.

Khan told us that access to the Maestro model will be available via a separate $20 (roughly Rs. 1,750) subscription tier. Initially, users will be able to generate unlimited soundtracks, but only download 15 minutes of generated output per month. The music generation model will also be provided as an enterprise offering to music studios and other players in the industry.

Beatoven.ai, as mentioned in the press release, has focused on using AI to generate music that does not mimic human sounds. Notably, its rival Suno AI is known for generating music and vocals that closely resemble human musicians.

Answering the reason behind taking a different route, Khan said, “We have found that creators and musicians prefer a co-creation tool, from a utility standpoint. Instead of giving them a readymade output, Maestro generates music that complements their creativity and vision for music. We want to enable that.”

Beatoven.ai is currently available as a website and a web app. However, Khan told us that the company is planning to launch a mobile app by the end of this year.