The audio software program Audacity recently introduced new effects AI OpenVINO of for users Windows. Among the new features announced there are useful tools, such as the noise suppression and the transcriptionuseful for podcast and the music generation. Users will also be able to leverage artificial intelligence for remixes of musical styles and the separation of music. The latter is useful for separating the various voices or instruments in a song. OpenVINO (Open Visual Inference e Neural network Optimization) is a toolkit released by Intel that allows local execution of deep learning models su hardware Intel.
Audacity: OpenVINO tools for editing audio and music using AI
Intel’s suite of AI tools, created for Audacity, is useful for both audio and music and works 100% locally on the PC. Going into detail, the noise suppressionuseful for podcasts, behaves similarly to the effect Noise removal, integrated into Audacity. There transcriptionprovided by Whisper.cppis able to transcribe, but also to translate words and output to a label track. To export the transcripts simply go to “File“, Then “Export more” and finally, “Export labels”. As for the AI tools for music, Music Generation e Music Style Remix they use Stable Diffusion (and Riffusion in particular) to generate new music respectively from a prompt o based on music pre-existing. Finally, musical separation can divide a song into its own vocal parts e instrumental or in vocals, drums, bass and a combined “nothing else” part.
To use the plugin OpenVINO, developed by Intel, you need to download them from the dedicated GitHub page. Audacity notes that only the Windows version is currently available for download. The project could be compiled on Linux and macOS, although no instructions are yet available for the latter.