No recording needed
Design a voice from a description
Write a few sentences about the voice you have in mind and VocaTTS generates an original synthetic voice to match. Save it to your account and use it anywhere you would pick a stock voice.
Design your voice
Sign in to create your personal voice
Personal voices are saved to your account and can be used in Text to Speech, Multi-Voice and SRT to Speech.
Sign inWhat to put in a voice description
- Who is speaking: age range and role (teacher, host, narrator, support agent)
- Accent or region, if it matters for your audience
- Timbre: deep, bright, breathy, crisp, gravelly
- Energy and pace: calm and slow, upbeat and quick
- The feeling listeners should get: reassured, excited, focused
Description examples
Online lessons
A warm, friendly female voice in her late twenties, clear articulation and a steady, unhurried pace, like a patient tutor.
Documentary
A mature male narrator around fifty, low and resonant, measured delivery with thoughtful pauses.
Product demo
A confident, upbeat voice in its thirties, crisp and modern, speaking at a lively but easy-to-follow pace.
Bedtime stories
A soft, gentle female voice, slightly breathy, slow and soothing, with a smile in the delivery.
If the first result is close but not right
- Change one trait at a time — age first, then accent, then timbre — so you can hear what each change does.
- Delete voices you will not use; each saved voice takes one slot on your plan.
- Keep a copy of descriptions you like so you can recreate a voice later.
Use a designed voice everywhere
Requirements and limits
- Description
- 20–1,500 characters
- Settings
- Voice name, gender (male or female) and main language
- Main languages
- 15, including Vietnamese, English (US/UK), Spanish, Portuguese (Brazil), French, German, Japanese, Korean, Chinese, Hindi, Indonesian, Thai and Italian
- Plan
- Basic plan or higher (3 custom voice slots; Premium and Studio: 5)
- Storage
- Kept for up to one year; delete it any time to free the slot
- Credits
- Speech generated with a designed voice uses the Gemini voice rate
Frequently Asked Questions
Q: Do I need to record anything for Voice Design?
A: No. A designed voice is generated entirely from your written description, so no microphone or sample is needed.
Q: Can I design a voice that sounds like a celebrity?
A: No. Voice Design creates original voices, and descriptions should not ask for an imitation of a specific real person.
Q: Which languages can a designed voice speak?
A: You pick one of 15 main languages when you create it, and the voice is built for that language. For another language, design a separate voice with that language selected.
Q: Where do designed voices appear?
A: Under “My voices” in the voice picker of Text to Speech, SRT to Audio and Multi-Speaker.
Q: What does a designed voice cost?
A: Creating one needs a Basic plan or higher and uses a custom voice slot. Speech generated with it is billed at the Gemini voice credit rate.
Already have the right voice in your head?
Write it down and hear it in a minute.