AI Audio Effects, Including Voice Transcription, Come to Audacity
Intel has released an array of AI effects for the open-source audio editor Audacity.
Intel’s OpenVINO AI plugins for Audacity (to give them their official name) have been designed to work offline and locally, meaning no cloud servers, signups, or subscriptions are required to use them — just some halfway decent hardware!
Include are time-saving effects for editing spoken word audio, such as podcasts, audiobooks, and interviews, as well as effects for music creation/production (the latter further bolstering Audacity’s growing DAW capabilities):
- Noise Suppression to, well, suppress unwanted noise(s)
- Transcription to transcribe and/or translate words & output to label track
- Music Generation to generate new music from a prompt
- Music Style Remix to generate new music based on pre-existing music
- Music Separation to split a song into vocal and instrumental parts
OpenVINO stands for ‘Open Visual Inference and Neural network Optimization’. It’s a new, open-source toolkit developed by Intel to run “AI inference” tasks “locally” on Intel hardware.
The Transcription effect is powered by Whisper.cpp, while the Music Generation and Music Style Remix use Stable Diffusion and Riffusion.
These new AI effects sound useful but there is a ‘bum note’ to using them on Linux: they have to be manually compiled from source code — both the OpenVINO AI effects and Audacity. No pre-built Linux packages are currently provided.
Helpfully, OpenVINO has a step-by-step guide that walks through the entire build process, start to finish. Better yet, it’s written against Ubuntu 22.04 LTS (though may work with later versions).
The Linux guide seems comprehensive though I’ve not tried it personally. Compiling requires a sea of dependencies (some several GB in size) which would subsume my patchy 3G data. Plus, given how painfully slow my laptop ran the GeekBench ML benchmark tool…
I’ll leave it to you 😉
Windows users are catered for, with pre-built packages containing all of the above-mentioned OpenVINO AI effects available to download from the OpenVINO GitHub releases page (though the effects do require the most recent version of Audacity for Windows to be installed).
Scoot over to the Audacity blog announcement for more details about the OpenVINO AI effects.