Feature Requests

Feature request for the ability to adjust the speed of generated Overdub audio. Looking for Controls to speed up / slow down the cadence. Ability to adjust the spacing between words and / or phrases.

I’ve been a user and fan of Descript for some time, especially the innovative AI tools that make video editing so efficient and accessible. I recently discovered that with the update to the pricing structure and the shift to legacy plans, the ability to create custom AI voices for text-to-speech was removed from the Hobbyist (Legacy) plan. This was a feature I actively used and one of the key reasons I chose Descript in the first place. It feels disappointing to see a feature I used taken away not because it was being retired, but seemingly as a push to upgrade to a newer paid plan. I understand the need to restructure pricing and introduce new plans, but removing existing functionality from legacy tiers feels like a penalty for long-time users. A legacy plan should preserve the value and features users originally signed up for, especially when they haven’t downgraded but were simply moved as part of an internal shift. I’m respectfully requesting that you consider restoring the custom AI voice feature for those of us on the Hobbyist (Legacy) plan. It would go a long way in maintaining trust, loyalty, and goodwill among your early adopters and continuing users.

It would be great to upload a video with narration and replace it with an AI voice while keeping the narration in sync with video. Specifically if we can pair this with translating the video into different languages.

I have tried two versions of Overdub: 1 with the classic training model, where I recorded 30 minutes of script; 1 with the new, faster training. The first option is FAR superior. Is there still a way to use the old training method to create a new speaker? We're interested in using Overdub, but this would be a dealbreaker.

Please increase the AI speech hours available per month in the creator plan. Currently, it only offers two hours. Alternatively, consider allowing users to purchase additional speech hours as an upgrade without having to switch to the business plan.

In my current projects with Avatar, there's no natural pause between the sentences it reads—the AI speaks continuously without a break. I’d prefer not to insert gaps or scenes between the current ones, which causes the screen to black out and creates a visually jarring result. It would be great to have audio blocks with export functionality and settings to adjust the reading speed slightly. Additionally, adding controls to insert natural pauses or "breathing spaces" between sentences would significantly improve the flow. Lastly, I'd like to have the ability to set the duration between scenes manually, so transitions feel more intentional and smooth.

I'd love you guys to train the stock voices to include Australian English. I've used Descript as a customer now to make training videos for our government here and they always ask to use Australian voices.

Descript now supports transcription in 23 languages, but Overdub only works in English. This feature would bring Overdub support to more languages