Convert text or a script to audio with our realistic speech features. Pick from many AI voices or build a custom voice clone in just a few minutes. Perfect for podcast intros, voiceovers, videos without on-screen hosts, and more.
Get started
These companies use Descript. Not bad!
01
Type or paste in your text
In a new Descript project, type or paste your script into the text editor, or use the Ask AI command in the Actions menu to draft text based on your chosen parameters.
02
Choose an AI voice or clone your own
Press ‘@’ to add a speaker to your script. You can create a new speaker name and then Enable speech generation to clone your voice. Or select Browse stock AI speakers to pick from a wide range of realistic options, including various styles and tones.
03
Generate your AI speech
Your script will display a brief loading icon while your audio is generated. After it’s ready, review your newly formed voice content, continue building an audio or video project, or export it by clicking Publish.
Generate and edit voice audio by typing
Descript enables you to turn your script to audio and edit by typing. Export the final result as MP3, WAV, or other common formats—all within one tool.
20+ realistic AI voices, emotions, and styles
Descript’s text-to-speech (TTS) features rely on advanced AI to create authentic voices. Pick from casual or formal tones to fit your project.
Create AI voice clones in minutes
Design and share personalized AI voices for ongoing work. Let AI manage voiceovers or subtle updates so you don’t need another recording session.
Descript is an AI-enhanced audio and video editing tool that helps you create podcasts or videos in a straightforward manner, much like editing text.
Captions & subtitles
Attach captions and subtitles to any text-to-speech project. This step supports accessibility for everyone.
Regenerate
Create a tailored voice clone to correct misreads or re-recorded lines with your original vocal character.
Podcasting
Produce, release, and share your audio or video podcast without complex steps.
Studio Sound
Enhance your audio by removing filler words and other issues for a polished final result.
Donna B.
Surely there’s one for you
Free
per person / month
Start your journey with text-based editing
1 media hour / month
100 AI credits / month
Export 720p, watermark-free
Limited use of Underlord, our agentic video co-editor and AI tools
Limited trial of AI Speech
Hobbyist
per person / month
1 person included
Elevate your projects, watermark-free
10 media hours / month
400 AI credits / month
Export 1080p, watermark-free
Access to Underlord, our AI video co-editor
AI tools including Studio Sound, Remove Filler Words, Create Clips, and more
AI Speech with custom voice clones and video regenerate
Most Popular
Creator
per person / month
Scale to a team of 3 (billed separately)
Unlock advanced AI-powered creativity
30 media hours / month
+5 bonus hours
800 AI credits / month
+500 bonus credits
Export 4k, watermark-free
Full access to Underlord, our AI video co-editor and 20+ more AI tools
Generate video with the latest AI models
Unlimited access to royalty-free stock media library
Access to top ups for more media hours and AI credits
Can someone else replicate my voice in Descript?
No. That can’t occur without your specific approval. Your voice data remains secure, and you can remove it at any point. We place user privacy first and follow our detailed code of ethics.
Can I use Descript's TTS generator for free?
You can make up to 5 minutes of text-to-speech audio at no charge. Once you pass that limit, upgrading gives you 120 minutes of TTS each month and other AI features, starting at $24/month.
Is there a difference between text to speech generated with a free subscription vs. a paid plan?
The free option offers 5 minutes of text-to-speech audio and 5 Regenerate actions. With a paid plan, you get higher monthly limits—for example, at $12/month, you receive 30 TTS minutes and 10 Regenerate uses, plus extra benefits.
How can I improve the quality of my text-to-speech voice clone?
Improve your text-to-speech voice clone by recording in a quiet spot, speaking clearly, and using reliable gear. Sticking to Descript’s recording suggestions in the prompt also helps create better outcomes.
