The new Overdub: Hear the difference

Record or import audio, make edits, add fades, music, and sound effects, then publish online, export the audio in the format of your choice or send it directly to your hosting service.
Create your podcast from start to finish with Descript.

On October 6 we released our newest version of Overdub — the magical voice-cloning tool that turn words you type into audio that sounds just like you — and it’s the biggest improvement we’ve made so far.  

If you’ve used Overdub in the past you’ve probably noticed a slight wobble in your Voice. It could be subtle, but it was one of those things that made it easy to tell when you were using a clone voice instead of your real voice. 

With the new version of Descript we’ve eliminated the wobble (“we” meaning our AI teams; the rest of us just watched). Overdub Voices sound far more natural and realistic than before — and they were already the most natural and realistic clone-voices out there. 

Here’s a before and after so you can hear the difference (you can hear it better with headphones): 

If you already created an Overdub Voice, or more than one (you can make as many as you want, as long as they’re all your voice), you’ll have to re-train it — but that just requires clicking the Update button in the Overdub section of the app. 

If you haven’t created one, you should. You’ll find it saves you all kinds of time and frustration — you can add or change words in your audio by just typing them, rather than trudging back to your mic to re-record. It’s free for all Descript users to try (only Pro accounts have an unlimited vocabulary) and all you need to do is upload or record at least 10 minutes of yourself speaking. 

Learn how to create your Overdub Voice in the Help Center. Or just go to the app, click Overdub, and follow the instructions — it’s easy. 

Featured articles:

No items found.

Articles you might find interesting

Product Updates

Here's what creators are saying about Descript Season 1: Overdub, Studio Sound + more

We asked creators to test out the latest features from Descript's Season 1 update, including our voice cloning Overdub software, Studio Sound audio enhancement tool and more.


How to go viral on TikTok: What works and what doesn't

There's lots of advice out there about how to go viral. A successful TikToker explains which strategies are actually effective — and which are just a mirage.


How to record a high quality podcast (the easiest way possible)

High-quality audio requires the right podcast equipment, solid recording techniques, and basic audio editing chops. Let’s explore what you need for your podcast.

Product Updates

Ultra Fast Audio Synthesis with MelGAN

In this post, we introduce MelGAN, a new generative model of raw audio waveforms created by the Lyrebird team that is capable of generating natural sounding speech at a rate of more than 2,500,000 audio samples per second — more than 100x faster than real time, and 10x faster than alternative methods on similar hardware.We believe that MelGAN paves the way for taking many real-time speech applications onto smaller devices. Imagine, for example, in the not too distant future, having real-time text-to-speech translation on your mobile device without the internet. And, it’s application to music translation brings us one step closer to AI-assisted music composing.We’ve open sourced MelGAN and we encourage interested machine learning developers and researchers to check out our code base.


Podcast metrics 101: Crunch the numbers to improve your show

Having hard data on your listeners’ habits can give you a sense of what’s working and what’s not — and help you set realistic goals for yourself and the show going forward.

Related articles:

Share this article