PrimeDictate 1.0.0 — First Release

English

Welcome to the first release of PrimeDictate.

PrimeDictate is a privacy-focused, system-wide dictation application for Windows. It records speech from a global hotkey, transcribes it with the engine you select, and safely delivers the result to your active workspace.

Highlights

• System-wide Toggle and Push-to-Talk dictation modes
• Local Whisper on CPU, NVIDIA CUDA, and AMD / Intel / NVIDIA Vulkan
• Optional cloud transcription with Groq, OpenAI, and Google Gemini
• Local rule-based cleanup or advanced text processing with a selected AI provider
• Turkish and English interface with multilingual speech recognition
• Long-form MP3, WAV, MP4, M4A, MKV, FLAC, and OGG transcription
• TXT, SRT, VTT, and JSON exports
• Movable floating dictation control and system tray model controls
• Optional muting of other Windows applications during recording
• Local dictation history and safe clipboard-based text delivery
• Windows startup and elevated-application support

Local CPU, CUDA, and Vulkan transcription keeps audio on the device. Cloud speech-to-text is used only when explicitly selected or permitted as a fallback. API credentials are protected with Windows Credential Manager or user-bound DPAPI storage.

Choose PrimeDictate-Setup.exe for the installed edition or PrimeDictate-Portable.exe for the standalone edition. Use SHA256SUMS.txt to verify your download.