Poor audio quality has always been a major pain point for video creators, from hobbyists filming with their phones to Hollywood professionals. Background noise, wind interference, and less-than-ideal mic positioning can wreck an otherwise great video.
Adobe is trying to fix this now with a new AI-powered solution called Project Sound Lift that stands out by actually letting users separate specific sounds from a clip. Let’s take a look at what it’s like and how it works.
What is Project Sound Lift?
Unveiled at Adobe’s Sneak Peeks event on November 2nd, Project Sound Lift is a new AI tool that can automatically separate the components of an audio track into distinct layers. Using advanced machine learning algorithms, Project Sound Lift analyses a complex audio recording and splits it into separate tracks for speech, background music or noise, applause, traffic sounds, and more – whatever your requirement may be.
This gives video creators granular control over each audio element. For instance, you can mute or adjust the volume of background noise while keeping the vocals crisp and clear. Or single out the sound of a piano from a band performance. The AI does the heavy lifting of isolation and separation.
How does it work?
Project Sound Lift integrates Adobe’s existing Enhance Speech technology, which is now available in Adobe applications like Premiere Pro, into a user-friendly one-click solution. After importing an audio file into the application, you have to choose what sound you want to filter out. There are multiple options to choose from, including applause, laughter, speech, typing, crowds, traffic, and more.
The AI will scan the file, detect distinct audio elements and generate separate tracks for each component. People have been using audio separation tools for a variety of use cases already but those often only let you tweak certain parts of the file or single out audio. Sound Lift goes a step further by allowing you to pick specific sounds from a clip.
What can you do with it?
The potential use cases for Project Sound Lift’s AI-powered audio separation are nearly endless. Here are just a few examples of how it can help video creators:
– For journalists, reduce background noise in interviews shot outdoors or in crowded settings
– Adjust levels on voiceovers or ASMR recordings to perfection
– Make vocals clearer for music videos or performances
– Amp up applause or reactions for event videos
– Isolate the audio from specific speakers in roundtable videos
Of course, these are only some of the scenarios we imagine people would put Sound Lift to use in. But as is often the case with innovative AI tools like this, their true potential is realised once they actually reach the devices of a large number of people.
When is it coming out?
Adobe has not announced an official launch date for Project Sound Lift yet. As one of their Sneak Peeks demos, it’s likely still in the development and polishing stages. However, the technology already looks very promising based on the initial previews.