Use a short reference clip to create a voice for your audiobook. Processing stays on your Windows PC and NVIDIA GPU.
Voice cloning basics
Upload a clear, single-speaker .wav or .mp3 clip, 3–30 seconds (8–15 recommended). Avoid music, echo, clipping and overlapping speech. Check the transcript, name the voice, and use it anywhere in the app.
Import an EPUB, choose chapters, assign voices, and export a chaptered .m4b.
Preview a voice before assigning it.
Use the same voice throughout a long book.
Remove voices you no longer need.
Design a voice from a description
Describe the sound, provide a line to speak, and compare the result with your other voices. Designed voices can be previewed and cast like cloned voices.
The 25 preset designs
Each design below is a ready‑to‑use clone voice. The app previews voices on your own GPU; no audio is hosted on this site.
Showing 25 of 25 preset designs.
Deep narrator male
Calm, confident, epic, middle‑aged.
Warm guide female
Early 30s, smooth American, reassuring.
Young energy male
Early 20s, bright, fast, playful.
Wise older female
70s, soft American, low and soothing, patient mentor.
Gruff adventurer male
Mid 40s, raspy, direct, battle‑worn.
Mysterious female
Slow, hushed, intriguing.
Epic villain male
Dark, menacing antagonist.
Cheerful young female
20s, bouncy, optimistic.
Calm neutral male
30s, steady, trustworthy.
Elegant female
30s, refined British, poised.
Grandpa storyteller male
70s, gentle, slightly hoarse, bedtime pace.
Dramatic female
Brings scenes to life.
Laid‑back Australian male
30s, relaxed, friendly.
Cozy Scottish female
40s, deep, melodic, slow.
Modern assistant male
Neutral, slightly synthetic, measured.
Passionate Italian female
30s, lively, musical.
Precise German male
40s, controlled, structured.
Southern drawl female
40s, soft and slow.
Polished English male
Mid 30s, standard English, articulate.
Gothic female
30s, dark, moody, low, British.
Tavern keeper male
40s, hearty, welcoming.
Ethereal angelic female
20s, airy, gentle.
Commanding leader male
40s, firm, strong.
Playful companion female
20s, mischievous, lively.
Neutral androgynous
30s, soft, gender‑neutral.
Browse 25 preset designs and preview them locally before casting.
Tips for a great reference recording
Use one relaxed speaker with clear diction.
Record in a quiet space without music, echo or clipping.
Keep the clip between 3 and 30 seconds; 8–15 seconds is recommended.