Aria
Northern Irish and American accents
Sits in a middle register.
The Mimicore voice gallery
Nineteen original voices that are free to use in Mimicore: live in Discord and games, in text to speech and in video dubbing. Each one is a blend of two speakers, so none of them copies a real person. Press play on any voice to hear a sample.
Female voices
Maya
American accents
Sits in a middle register.
Nova
American and English accents
Sits in a middle register.
Iris
English accents
Sits in a middle register.
Luna
English and American accents
Sits in a lower-middle register.
Zara
American accents
Sits in a lower-middle register.
Elena
American and English accents
Sits in a middle register.
Kiara
English accents
Sits in a higher register.
Sage
English and American accents
Sits in a higher register.
Rowan
American and Irish accents
Sits in a lower-middle register.
Male voices
Leo
American accents blended with Mimicore's own default voice
Sits in a middle register.
Kai
Scottish and American accents
Sits in a lighter register.
Owen
American and Indian accents
Sits in a middle register.
Rafael
Indian and Australian accents
Sits in a middle register.
Theo
Australian and English accents
Sits in a lighter register.
Marcus
English and Canadian accents
Sits in a middle register.
Ezra
Canadian and English accents
Sits in a deeper register.
Felix
English and American accents
Sits in a deeper register.
Jonah
American and English accents
Sits in a deeper register.
How to use a gallery voice
- Install Mimicore and its Instant voice cloning add-on.
- Open Voices & effects, find the voice in the Voice gallery and press Add to my voices.
- Pick it on the Live page and press Start, or choose it in Speak and Dub.
Gallery voices work live without any limit, on the Free plan too. You can also blend any two voices into a new one with the mixer, and clone your own voice from about 15 to 20 seconds of speech. See the how-to guides to use the voice in Discord, OBS, Zoom or a game.
Voices are blended from speakers of the CSTR VCTK Corpus (CC BY 4.0, Centre for Speech Technology Research, University of Edinburgh) and Mimicore's default voice. No gallery voice reproduces a single real person.