Using AI to Improve a Short Informational Video
As technology continues to advance, it is becoming increasingly common to incorporate AI tools into various aspects of our lives. One area where AI is proving to be particularly useful is in the field of video production. In this article, we will explore how one YouTuber used three AI tools to improve a short informational video for his channel.
The project began with a problem the YouTuber encountered with his Glowforge, a laser cutter used in many maker projects. He was using it to make acrylic two-color labels for special adapters he 3D-printed, but the resulting laser cut was the wrong size. To document his solution and share it with other Glowforge users, he decided to create a quick informational video. However, due to time constraints, he did not put much effort into the production design, and the audio quality was poor.
The first AI tool the YouTuber used was Adobe Podcast, which is an experimental AI audio tweaking tool. He used it to clean up the poor audio quality in the video. To do this, he started with the original MP4 video recorded on his iPad and used QuickTime to export the audio track as an M4A file. Unfortunately, Adobe Podcast does not support M4A, so he used Adobe Audition to open the M4A and save it back out as a WAV file. He then uploaded the audio track to Adobe Podcast and used the Enhance Speech option to improve the audio quality significantly.
The second AI tool the YouTuber used was the Scene Removal Mask in the latest version of Apple’s Final Cut Pro for Mac. This AI-powered masking tool provides green screen functionality without the need for a green screen. To use it, the YouTuber captured a second or two of video with just the background (without the subject in the frame) and then captured the normal video. He applied the Scene Removal Mask to the video, which allowed him to drop out the background from behind the speaker and put anything behind the person in the video. This tool helped the YouTuber fix issues with the background and allowed him to enlarge himself to better fill the frame.
The third AI tool the YouTuber used was Adobe Generative Fill in the Photoshop Beta. This tool allowed him to clean up the background of the video. He used it to remove a 3D printer that was blocking the corner of the Glowforge and draw in the missing area of the wall. What amazed him was that the AI drew in the rest of the Glowforge, including the texture and lines, that were originally missing from the image. He also added a Daleks poster from Doctor Who into the background using Photoshop’s non-AI distort feature with a blend mode.
Overall, the YouTuber found that using AI tools significantly improved the quality of his video and saved him a lot of time. Without the help of AI, he may not have bothered to fix the poor audio quality or put in the effort to improve the production design. As AI technology continues to advance, it will be interesting to see how it can be used to improve video production even further.
