|
iZotope has launched Photon, a standalone app and ARA plugin for AI-powered stem separation and audio enhancement. They say that it can can split music into four stems, separate drums into individual elements, isolate vocals, identify and separate speakers, and clean up noisy recordings. Here's the details direct from the company...
Powered by 12 cutting-edge custom-trained models, Photon runs locally on the GPU for fast processing without any uploads. Whether you're cooking up a remix, adjusting the kick volume, exporting a quick acapella, untangling multiple speakers in a podcast or interview, or cleaning up spoken word, Photon delivers. Choose your model, let the magic happen, export, and you're done. Or use Photon as an ARA plugin in your DAW: perfect for making quick balance upgrades to a stereo stem. Photon is the first collaboration between the minds at iZotope and Boris FX, bringing together two world-class research teams to create cutting-edge machine-learning models that sound incredible.
Included machine-learning models
Demix -- split the mix
• Music - 4 Stems Ultra: new 4-stem high-quality stem separation model for music.
• Drums - 5 Stems: magically adjust the kick, snare, hi-hat, toms, and cymbals.
• Voice Separator: split up to 6 speakers from one audio file into individual stems.
• Voice - 2 Stems: voice and music, for a quick instrumental or acapella.
• Cinema - 4 Stems: split a video or audio file into dialogue, music, sound effects, and everything else.Denoise -- fast and powerful cleanup
• Voice Restore: enhances speech and handles noise, levels, mic tone, and reverb.
• Indoor / Indoor Noisy: tames reverb and room noise, enhancing dialogue to production-ready, fast.
• Outdoor / Outdoor Traffic / Outdoor Wind: optimized models for traffic, urban noise, and heavy wind.Additional features
• ARA plugin: Tweak stems or fix background noise right in your DAW. Run several models together and switch between them without waiting.
• Local GPU processing: large, custom-trained models run locally on your GPU for fast, private processing without uploads.
• Preview from anywhere: start processing at any point in the clip and hear results in seconds. Very helpful for long files.
• Model Manager: download new and updated models without needing to reinstall.
• Flexible exports: deliver in WAV, FLAC, MP3, or video same-as-source with the original video stream kept.
Pricing and Availability:
$349 MSRP. $259 early cyber deal through December 7, with extra loyalty deals for existing customers.
More information:
Trawling our Legacy Site of Candid Synth Opinions