Clone a voice, then
direct it
Fifteen seconds of clean audio builds the model. Write the script, render the take, hear it back.
15-90 s of one speaker
1 credit = 1 minute
Charged by the second
Append-only ledger
Script and takes, one screen
write it. render it. hear it back.
One screen, three moves. The take comes back at the length you asked for and the ledger settles against what actually rendered.
Built to be worked in, not called
The transport, the script and the take sheet sit on one screen. You write a line, render it, and hear it back without leaving the room.
Fifteen seconds is the whole capture.
One speaker, recorded with a live meter or dropped in as a file. Past ninety seconds you are not building more model, you are waiting longer.
A voice is a model, not a recording.
Rewrite the line and render it again at a different length without going back to the speaker. The sample it was built from stays attached to it.
Every model, and the sample it came from
Explore the Resona voice selection. Press play on any row to hear the exact sample that model was built from.
- 01
Sarah
Englishpublic modelsample
- 02
Adrian
Englishpublic modelsample
- 03
Jade
Englishpublic modelsample
- 04
Farid Dieck
Spanishpublic modelsample
- 05
Sophie
Frenchpublic modelsample
- 06
ふうか
Japanesepublic modelsample