HushVoice Studio
Engine offlineStart with docker compose up -d engine

Step 1

Record the sentence

Read the prompt aloud in a quiet place. Aim for 8–12 seconds of clear speech.

“The quick brown fox jumps over the lazy dog near the riverbank at sunrise. Clear voices carry farther when spoken with calm confidence.”
0:00·target 0:12
x

Step 3

Type anything

HushVoice synthesizes speech in your cloned voice. Switch languages anytime — models run on your server.

59/2000 · No voice selected yet