Overdub is finally here. This state-of-the-art voice generator allows you to create an ultra-realistic clone of your own voice — so you can create speech just by typing text in Descript.
Audio has always been the simplest form of content to create — you simply record yourself speaking — but the hardest and most time-consuming to edit. Overdub changes that. If you want to change an adjective, add additional context, or reword a sentence in the middle of a podcast or video, you no longer need to make a trip to the recording booth to do so.
Want to easily begin creating lifelike text-to-speech in your own voice? Read on.
If you have a decent microphone and ten minutes to record an Overdub training script, that’s all you need to start using Overdub. To train the software with your own voice, you’ll need to record yourself speaking following a script that we provide.
10 minutes is the minimum length necessary to create your Overdub Voice, but to ensure production-quality results, we recommend recording 30-90 minutes of training data.
Perhaps you recorded an hour-long podcast or training video, but wish you had worded a sentence differently or forgot to mention an important detail. Find the right sentence in your transcript, delete what you wish to remove, and type in what you’d like to add. Overdub will make the change for you in your own voice — no recording necessary!
When it comes to scratch tracks for animation, video games, scripted audio or videos, and beyond, Overdub will save you hours of work and time. You might have thousands of lines of audio to work through, but you can use Overdub to easily check pacing and timing of new or proposed content. Overdub will help you test, refine, and re-test material without having to record and re-record.
Overdub also works wonders for voiceover and dialogue. Overdub users can even share access to their Overdub Voice with other Descript users, allowing team members to easily add speech to a Composition in a voice other than their own.
With the right setup, Overdub can be indistinguishable from real audio. The quality of your particular Overdub Voice will depend on several factors:
Please read the Overdub Guide to Voice Quality for a comprehensive set of guidelines.
No. You may only create an Overdub Voice using your own voice. As part of the Overdub Voice training process, users must read a script in which they affirm their identity and positively consent to Overdub Voice creation.
Training data which does not include this consent statement cannot be used to create an Overdub Voice. In other words, unless someone specifically and verbally consents to Overdub Voice creation, their voice cannot be used to train an Overdub Voice.
Security, privacy, and confidentiality are essential priorities for Descript. Read the Overdub section of our Security and Confidentiality policies for more details.
Yes. When you’re creating a standalone Overdub, or creating a sentence from scratch, you’ll be able to select from different vocal styles and intonation for the sentence you generate.
When you’re editing a sentence you’ve already recorded, Overdub will automatically match the vocal style of the sentence you’re editing.
To get started using Overdub, Descript users will need a Descript Pro account. Every Descript Pro subscription includes a 7-day free trial, so you can test Overdub and make sure it meets your needs.
Download Descript now and try Overdub for yourself.