❓ Knowledge Base & Assistance
Frequently Asked Questions
Find comprehensive answers regarding translation accuracy, technology standards, supported idioms, data protection, and voice features.
State-of-the-art AI translation achieves 95-98% accuracy on general, technical, and business texts by understanding full-paragraph context and cultural idioms. While human post-editing is still recommended for critical legal certifications, AI delivers instant, natural, and low-cost translation for everyday communication, research, and documentation.
We support over 50 global languages including English, Spanish, French, German, Mandarin Chinese, Japanese, Korean, Arabic, Urdu, Hindi, Portuguese, Russian, Italian, Dutch, Turkish, Vietnamese, and many others, complete with auto-detection.
Speech-to-Text uses the standard W3C Web Speech API built directly into modern web browsers. It accesses your microphone locally to transcribe audio into text in real-time, which is immediately translated by our neural engine.
Yes. Custom Gemini or OpenAI API keys are stored exclusively in your browser's private localStorage sandbox. They are never transmitted to our backend servers or logged in any database.
Yes! The application features a lightweight, zero-dependency client architecture and a built-in deterministic offline heuristic dictionary that provides instant fallback translations even during intermittent network disruptions.
For languages such as Arabic, Urdu, Persian, and Hebrew, our styling automatically adjusts text directionality (dir="rtl") to ensure punctuation, numerals, and sentence flows render properly.