Descript Has Everything You Need to Edit Audio and Video Easily

Descript is an online service with several tools to help you edit audio and video files painlessly. There is a free version that is limited, but if you need more transcription time or a larger vocabulary for your voice clone, there are three different paid tiers.

Overview Video

If you would prefer to watch an intro video created by Descript, you can view it here:

Key Features

  • Use Descript to capture your screen and record your microphone or computer audio.
  • Transcribe your audio or video at the press of a button.
  • Record remotely
  • For podcasting: edit audio, remove silence, add crossfaces and effects.
  • Edit video, add titles, shapes, lines, arrows, and images.
  • Remove uh, um, and other filler words instantly.
  • Overdub: create a digital clone of your own voice to generate and edit audio tracks.

Overdub

The feature that really made me stand up and take notice of Descript is Overdub. It allows you to create a clone of your voice (free version limited to 1000 words). You can then create an audio track of your voice by just typing the words! It can also be used to make changes to an existing audio recording and blend the tone on each side to make it sound natural. You can also create voices in different tones and performance styles in order to apply Overdub in a variety of situations.

While this technology has been around for some time, consumer tools have left a lot to be desired. Overdub marks a giant step forward in quality, with its AI doing the heavy lifting.

To hear samples of Overdub voices, check out this page: https://www.descript.com/overdub. Even better – if you’d like to take it for a live test drive, there is a widget at the bottom of the page that invites you to choose a voice profile and type in any text you want. Click the “speak it” button to hear your text “read” by the AI voice profile. Just below the widget, there is an option to test it against other popular text-to-speech services.

Conclusion

Descript is a free, easy-to-use tool that is full of features and can help you create and edit audio and video tracks easily. Have you used Descript or Overdub? Do you have another AI tool you find indispensable? Let us know in the comments.

Enlarge Images Easily Without Quality Loss

Have you ever tried to use an image online that looked pixelated or otherwise low-quality when enlarged? Most of the buzz in using online images is how to make the file sizes of the images smaller, but that often results in a quality loss that can leave you disappointed. The good news is that there are several AI-backed tools available to assist you in enlarging your images without losing quality.

There are a number of tools that can accomplish this, such as:

They all work similarly, though their free plans have different limitations. For the purposes of this demo, we are going to use https://bigjpg.com.

Upscaling an Image

The main page contains a standard upload window that allows you to choose an image on your device to upscale.

Screen shot of the home page with a "select images" button in the center

When your image has finished uploading, select “start.”

Screenshot showing the uploaded image with buttons to start or delete

The configuration window will appear. First, select the type of image you are using. Then, pick a factor by which to increase the image. For this tool, the free version is capped at 4x.

Screen shot of the configuration window with options for image type (artwork or photo), upscaling (2x, 4x, 8x - paid, and 16x - paid), and noise reduction (none, low, medium, high, highest)

In testing, I tried a few different images. I found that photos best showed the power of the noise reduction aspect of the tool. For this, I used one of our stock photos of a young man at the library. Here is the original:

Man in his 20s sitting between bookshelves with a stack of books

First, I enlarged the image 4x, and didn’t apply any noise reduction. Click/tap on the image below to make it full size. Depending on the size of your screen, you may be able to see how fuzzy the image looks in his face and his shoulders. That’s the “noise.”

Same image, but larger and lower quality.

I started from the original image and enlarged it 4x again, this time also opting for the highest noise reduction. Click/tap on the image below to make it full size, and note that this image has all of the “noise” smoothed out to make the picture quality appear better.

Same image at 4x, but the rough edges from before are gone

You may wonder why they offer different levels of noise reduction. To see the answer, compare the last image with the original. Where the AI has reduced the noise, it has also “guessed” at how to edit the image and left it looking a bit like a soft filter has been applied. In some cases, you may want to balance that effect with the noise for a better outcome.

Summary

Now that AI technology has become available to the consumer market, free image editing tools have gotten much more sophisticated. There are several sites that will allow you to upscale and enhance your images without a pricey software product. Have you used any of the tools we listed? Do you have another image upscaling tool you love? Let us know about it in the comments.

Avatarify: Make that Picture Sing!

Today we are highlighting Avatarify, a face animator app backed by artificial intelligence that is able to take an image and a song and make the person (or animal!) in the image sing that song. Then you can share the video with friends. The app is free and available from the Google Play Store or the Apple App Store. Keep in mind that while this is called a “deep fake” video, you aren’t likely to fool anyone with the resulting video.

Using the App

After installing the app on your device, the initial screen makes it appear that a subscription is required to use the app. Note that it is easily dismissed by tapping the x in the corner of the screen.

Pick a song from the Avatarify library.

Screenshot of the phone screen showing thumbnail images of songs in the library.

Next, tap “add photo” to select an image from your gallery.

Screenshot of the preview showing selected song with "add photo" and "continue" buttons under it.

The first time, you’ll be prompted to enable access to your gallery. This is required unless you want to use a demo image. For the first round, I chose a picture of my kittens. No matter how hard I tried, I couldn’t make the mouth land in the right place, giving the cat a monstrous look in the video. I gave up and tried a demo image of Elon Musk.

If you’re using your own image, it might detect the face(s) automatically. If not, the next screen shows a shadow face over the image (the part I had trouble with in the cat picture). Move it over a face and save. You can select multiple faces in the image to animate.

Screenshot of the shadow face on a cat image, showing the features do not line up.

When I didn’t have to aim the “shadow face” over the image, it was really easy! This is how mine turned out:

The quality of the initial video was higher, but I compressed the file to lighten the file load on this page.

After the video is created, you have the option to save it to your device or share it using any method available on your phone.

Would You Use It?

This was a fun little tool, and the demos on their webpage suggest a variety of use cases: https://avatarify.ai/. Is this a tool you would use to share videos with your friends? Let us know in the comments.