Whispo is an AI-powered dictation tool designed for seamless integration with any application supporting text input. It targets users seeking an efficient, local-first transcription solution that leverages advanced AI models for both speech-to-text and post-processing.

How It Works

Whispo utilizes a simple hotkey mechanism (hold Ctrl to record, release to transcribe) to capture audio. The core transcription is handled by OpenAI's Whisper model, with the flexibility to use custom API endpoints like Groq. Transcripts are automatically inserted into the active application. A key feature is its support for post-processing transcripts using Large Language Models (LLMs) such as OpenAI, Groq, and Gemini, enabling advanced text refinement.

Quick Start & Requirements

• Install: Download pre-built releases from https://github.com/egoist/whispo/releases.
• Prerequisites: macOS (Apple Silicon) and Windows x64 are currently supported. Requires access to OpenAI Whisper or Groq API.
• Setup: Local data storage.

Highlighted Details

• AI-powered dictation with automatic text insertion into any application.
• Supports OpenAI Whisper and Groq for transcription.
• Enables post-processing of transcripts with LLMs (OpenAI, Groq, Gemini).
• Data is stored locally for privacy.

Maintenance & Community

• Project maintained by egoist.
• No explicit community links (Discord/Slack) or roadmap provided in the README.

Licensing & Compatibility

• License: AGPL-3.0.
• Compatibility: The AGPL-3.0 license may impose copyleft restrictions on derivative works and linked applications, potentially impacting commercial or closed-source use.

Limitations & Caveats

Currently in preview, with builds only available for macOS (Apple Silicon) and Windows x64. The AGPL-3.0 license requires careful consideration for integration into proprietary software.