Adobe recently released a free audio processing tool that can make voice recordings sound better by removing background noise. It sounds like a recording made in a sound booth with a high-quality microphone.
The Enhance Speech tool was developed as a result of the Project Shasta. Project Shasta was renamed to AdobePodcast recently.
Enhance Speech can be used for free, but you have to create an Adobe account to use it. Users can upload an mp3 or WAV file up to one hour long or 1gigabyte. You can either listen to the result in your browser or download the cleaned-up audio.
Enhance Speech worked best with audio that contained a voice without cross talk or excessive noise. We recorded audio from an iMac's built-in microphone of a person standing 10 feet away, including fan noise nearby, and the resulting audio sounded like it had been recorded in a noise-free studio with a professional microphone.
Advertisement
Is it possible that it works? We think Adobe trained a deep learning model on thousands of hours of clean and loud audio. The model could learn to pick out human voices and make a facsimile that matches the source. We reached out to the company for comment until Adobe gives more technical details.
Some Hacker News commenters have reported hallucinated results, such as phantom voices where the artificial intelligence misinterprets the input audio, which suggests that Enhance Speech is doing more.
Enhance Speech is not the first tool to offer this kind of noise reduction. A commercial service called Audo Studio and an open source package called mayavoz do the same thing.
Enhance Speech is part of a larger group of podcasting tools from Adobe, including a mic check tool and transcript-based audio editing tool that is still undergoing an invitation-only alpha test.
All Rights Reserved © 2024
