Intel releases AI tools with music generation and transcription for Audacity

Intel has released an AI tool suite called OpenVINO AI for the open source audio editing program Audacity. The plugins allow users to locally transcribe podcasts and generate music based on prompts.

OpenVINO AI is only compatible with the 64bit version Audacity 3.4.2 for Windows and must be installed manually from files provided via GitHub have been made available. The repositories were already published on the platform by Intel a few weeks ago, but the announcing the tools by the organization behind Audacity only took place on Tuesday. For now, users on Linux or macOS will have to compile themselves. It is not clear whether the model will eventually be released natively for those operating systems.

According to the developers of Audacity, several plug-ins have been made available, including three tools for music, namely Music Generation, Music Style Remix and Music Separation. The first two tools use the Riffusion model, a branch of Stable Diffusion, to generate or remix music from text prompts. The Music Separation tool allows users to split an existing song into several tracks. According to the organization, the tool can split a song into vocals and instruments, or into the four tracks of vocals, drums and bass and remaining instruments.

Furthermore, two tools for spoken audio content have been made available: Noise Suppression and Transcription. These plugins allow users to remove background noise and transcribe spoken audio with AI, respectively. All OpenVINO AI tools run locally and therefore do not require an internet connection to Intel servers.