Frequently Asked Questions

Built-in dictation sends your voice to cloud servers. VoSonar runs AI models locally on your device — no internet needed, zero privacy leaks. It also understands context better and produces cleaner, more natural text with automatic formatting.

Yes. All speech recognition runs on-device. You can use VoSonar on a plane, in a basement, or anywhere with no connectivity. The only time VoSonar connects to the internet is to verify your license and download updates.

VoSonar supports 100+ languages and dialects with automatic detection. Major languages include English, Chinese (Simplified & Traditional), Japanese, Korean, French, German, Spanish, Portuguese, Arabic, and many more. The Free plan includes 2 languages; paid plans unlock all languages.

Yes, the Premium plan includes an advanced meeting mode that can distinguish between speakers (speaker diarization) and label each person's contributions in the transcript. For best results, use a good quality microphone.

Never. VoSonar processes all audio locally. We do not collect, store, or use your voice data for any purpose. Our business model is subscription-based — we sell software, not data.

Windows 10/11 (x64), macOS 12+ (Apple Silicon or Intel), or Ubuntu 22.04+/Debian 12+ (x64). VoSonar is lightweight — typically using under 500MB RAM and minimal CPU when idle. An SSD is recommended for faster model loading.