StoryVox Voice Cloner is a companion tool for StoryVox Pro that lets you create custom AI narrator voices from a short audio sample — trained entirely on your own PC, with no cloud processing, no internet connection, and no subscription. Record or import 10 to 30 seconds of clean audio from any speaker, drop it into Voice Cloner, and train a personalized voice style using GPU-accelerated local AI. The resulting voice file installs directly into StoryVox Pro, where it becomes a full narrator available for any audiobook project. TRAIN YOUR OWN NARRATORS Feed Voice Cloner a short clip — your own voice, a client's voice, or any voice you have rights to use. The local training engine extracts the vocal character into a compact JSON file. Training typically completes in under 30 minutes on a compatible GPU. VOICE LIBRARY MANAGEMENT Browse, rename, and delete your saved voice files directly from the app. Your voice library lives in your Documents folder — easy to back up, share, or transfer to another machine. PREVIEW BEFORE YOU SAVE Synthesize a short audio preview of your trained voice inside Voice Cloner before committing it to your library. Hear exactly what you'll get in StoryVox Pro before saving. SEAMLESS STORYVOX PRO INTEGRATION Saved voices import into StoryVox Pro with two clicks — File → Voice Settings → Import Voice. Your cloned narrator becomes immediately available for any project alongside StoryVox Pro's built-in voices. GPU REQUIRED FOR TRAINING Voice training requires an NVIDIA GPU with 8 GB or more of VRAM. Voice Cloner detects your hardware on launch and switches to browse-only mode if no compatible GPU is found — so you can still manage and preview your existing voice library on any Windows PC. 100% OFFLINE No internet required at any point. Your audio samples and trained voice files never leave your computer. No account needed, no usage monitoring, complete privacy. TRAINING TIPS Use clean audio recorded in a quiet room — no background noise, no reverb. Dense, natural speech trains better than dramatic pauses. Female voices typically clone with higher accuracy than male voices. See Help → How to Improve Your Voice Model for full guidance. System Requirements - Windows 10 or Windows 11 (64-bit). - Training: NVIDIA GPU with 8 GB or more VRAM required. - Browse and manage existing voices: any Windows 10 or 11 PC. - Minimum: Intel Core i5 or equivalent, 8 GB RAM. - No internet required after installation. * REQUIRES STORYVOX PRO Voice Cloner is a companion to StoryVox Pro. Trained voices are designed to be used with StoryVox Pro's audiobook production engine. Voice Cloner requires StoryVox Pro specifically and does not work with the standard StoryVox app