Content creators that deal with videos and audio know the struggle. The struggle to get good audio is universal for almost any modern content creator. In most cases, while most might not realise it, bad audio can break your immersion or a good video. Think about it, imagine watching The Conjuring on mute. You will find that it is way less scary than it is.
Good audio is not just limited to the music, or sound effects though. In the world of radio, YouTube personalities, and podcasts, great audio usually means great vocals. We understand that struggle because we produce our very own podcast (available on Spotify, coming back in 2023). We had to spend a little bit of money to get a set of gears for the office and home ready and going for our Podcast series, Tech & Tonic, just to get that crisp and smooth radio voice.
Thing is, we do have to spend a bit of our own money to get better audio even at home. Getting at least a dedicated USB mic was the only way to get quality audio for our podcasts during the pandemic. But what if you do not or do not have that money to spend on a mic? What if your regular built-in mic on your laptop or headphone is enough to create your own podcasts? It is possible today with Adobe’s latest free tool.
Adobe, as one of the most widely used content creation suite, has been researching on ways to improve creative workflow using AI for years. Their AI tool has helped make Photoshop, Illustrator, Lightroom, Premiere Pro, and even Rush even more efficient and way cleverer than before. They have also been messing about with AI in audio processing with Project Sasha, recently rebranded to Adobe Podcast.
As part of the project, they came up with something called Enhance Speech. The tool is free at this time, only requiring you to have an Adobe account to begin with, even without any subscription plans. The tool also works best with a desktop web browser, so you do want to get on at least a Chromebook to get it working. MP3 and WAV file format works best with the tool currently and the upload limit is 1GB.
Once you do that, the tool uses some form of deep learning algorithm to transform your audio from a slightly noisy recording to sounding like its recorded in an isolated professional studio. There have been reports that the tool does not work too well in extremely noisy scenarios though, so you do want to ensure that you are still doing as much as you can to keep your recording environment as quiet as possible. Of course, description only goes so far. You ought to experience the tool yourself, since it is free.
Our sample is a voice recording using the Samsung Galaxy Note9, one recorded with the mic right up near the mouth, and one placed on the table. We did not make too much effort in isolating the recording environment. We have some traffic noise, construction nearby, and even the MRT going about in our recording environment.
You can hear that a lot of the pop peaks have been cancelled out. While there is not much in terms of noise, the resulting audio sounds much cleaner and contains some of those radio like quality in the voice. Wind noise and echo is reduced to the point where you do not hear them too. If you are looking to start a podcast or improve your audio without spending extra money, you might want to give this tool a whirl on their website.