Whisper-Input-Next

An intelligent voice transcription input tool supporting multiple transcription services including OpenAI GPT-4o, GROQ, and local Whisper models.

Whisper-Input-Next is an enhanced voice transcription tool that brings powerful AI-powered speech recognition directly to your workflow. This project extends the original Whisper-Input with extensive feature expansions and architectural optimizations.

Key Features

  • Multi-platform Transcription Services: Support for OpenAI GPT-4o transcribe, GROQ, SiliconFlow, and local whisper.cpp
  • Smart Hotkeys: Quick access with Ctrl+F (OpenAI high-quality) / Ctrl+I (local cost-saving mode)
  • Audio Archive: Automatically save all recordings with history playback support
  • 180s Long Audio Support: Handle up to 3 minutes of continuous recording
  • Dual Processor Architecture: OpenAI and local processors working simultaneously
  • Privacy Protection: Local processing option ensures data security

Technical Highlights

The tool features a robust dual-processor architecture that allows seamless switching between cloud-based high-quality transcription and local cost-effective processing. With intelligent error handling, retry mechanisms, and a clean status indicator system, it provides a smooth user experience without polluting your system clipboard.

Visit the GitHub repository for more details and installation instructions.