As you’re probably well aware, there is a lot of talk about AI lately—for good reason. It’s one of the hottest topics in the world. We’ve got AI chatbots that can write love letters and AI that can give questionable advice. There is AI that can create images from text, and now AI can churn out some fairly decent videos. There’s chatter about how AI is being trained, and if companies should be able to use our content to do it.
While we aren’t quite to a place where we need to swear fealty to our robot overlords, AI has come a long way. There is a lot of fear about what AI means for our future in terms of possible jobs lost in multiple industries. Some believe AI might be the key to saving humanity. Whatever you think of AI, it’s important to remember this: most current AI available is little more than useful tools to be used by living and breathing humans to turn formerly annoyingly tedious tasks into far less annoyingly tedious tasks.
Today we’re going to highlight some AI tech Mad Genius is currently using in the world of video editing, specifically, the audio part.
Sound Design: An Unsung Hero
Sound is one of those underrated and unappreciated aspects of production. Sound editing is one of the most complicated tasks in the post-production process, as well as wildly time-consuming. It’s about sweetening the audio you want via EQ enhancement and noise reduction to the audio you don’t. There’s a lot of testing, listening, tweaking, and listening, and trying again. Thanks to Adobe Enhance Speech, cleaning up audio has been streamlined and much, much easier.
Adobe Enhance Speech is a feature currently available within Adobe Premiere Pro. But much like Diana Ross when she was a part of The Supremes, or Beyoncé in Destiny’s Child for our younger viewers, it’s going solo and will be sold separately. It’s designed to improve the clarity of speech in audio recordings, using some pretty advanced algorithms (that’s where the AI magic happens) to analyze the audio for speech specifically.
Using Adobe Premiere Pro’s AI Algorithms
Sound Without Processing
Let’s take a look, or rather a listen, to the tool in action. If you’re able, it’d be best to pop on a pair of headphones for the full effect. This first clip is without any audio processing.
Sound With Some Processing
You’ll notice that there is what sounds like a lot of hissing or white noise in the background of the voiceover. It’s the worst you’ll ever listen to, but it’s still clearly not something you’d want to hear. Now let’s listen to the audio enhanced via Adobe Enhance Speech, with a mixture of 55 percent process and 45 percent original audio.
Sound With All the Processing
The difference is quite noticeable. There’s far less hiss in the background, making the voiceover sound much sharper and clear. Now let’s listen to one final clip set to 100 percent processing.
The background noise is now completely gone, which can be what you want in some cases. The previous sample is actually what we ended up using because of its balance of minimum background noise or tone in relation to the voice-over.
That’s just one example of how AI is working for Mad Genius. If you’re looking to learn more about what we do and how we can help you and your business, why not fill out the contact form below and let’s chat.