Openai Is Reportedly Developing An Ai Music Generator Based On Prompts

The startup is partnering with Juilliard students to annotate musical pieces to enhance the AI's training data.

Jaque Silva/ Nurphoto/ Getty Images

OpenAI is allegedly creating an AI music-generation tool that converts text and audio prompts into whole tunes. Initial applications encompass evaluating movies and augmenting vocal performances with accompaniment. Numerous sources additionally report on trial collaborations with Juilliard students who are annotating scores to enhance training data.

No formal confirmation or launch date exists. The tool may be integrated within ChatGPT or connected to OpenAI's rapidly expanding video infrastructure, which already emphasizes audio. Sora 2, leading OpenAI's initiative in film and audio production, makes the development of a music generator a rational subsequent step; nonetheless, it raises significant concerns over licensing, consent, and revenue distribution.

The industry environment is strained. Leaders in AI music technology are confronting legal actions about their training data. Should OpenAI reach the market, anticipate pressure to finalize licensing agreements and implement safeguards to prevent style imitation. A refined tool for producers may provide immediate, customized cues, advertising jingles, and demo tracks without the need for costly libraries or studio sessions. For artists, it exacerbates discussions around attribution, remuneration, and the boundaries of imitation.

If implemented, this initiative would integrate music into OpenAI's comprehensive creative pipeline encompassing text, images, video, and sound, thereby elevating the stakes for the entire generative arts ecosystem.

OpenAI possesses a proven history in music artificial intelligence. The company previously released MuseNet (2019) and Jukebox (2020) — a neural network that produces music, including basic vocalization.