Amical provides a local-first, open-source AI dictation and note-taking application designed for enhanced productivity and privacy. It targets users seeking faster input methods than traditional typing, developers needing efficient note-taking, and anyone prioritizing data security, offering offline functionality powered by local open-source models. The core benefit is achieving significantly faster input speeds with intelligent, context-aware processing, all without sending data off-device.

How It Works

Amical leverages a robust tech stack including Whisper for accurate speech-to-text conversion and Ollama for running open-source Large Language Models (LLMs) locally. Built with Typescript, Electron, and Next.js, it operates entirely on the user's machine, ensuring privacy and offline capability. Its key innovation is context-aware dictation, which intelligently adapts formatting and understanding based on the active application (e.g., IDE, chat client, email), streamlining workflows.

Quick Start & Requirements

Installation involves an in-app, one-click setup for local models, simplifying the initial configuration. Specific hardware requirements (e.g., GPU, CUDA) are not detailed, but the local-first nature suggests sufficient local resources are needed for model execution. Links for "Download", "Website", "Docs", and "Community" are mentioned in the repository header.

Highlighted Details

• Super-fast dictation powered by AI-enhanced accuracy.
• Context-aware speech-to-text adapts to the active application.
• Privacy-first design with offline operation and local model execution.
• Extensible through hotkeys, voice macros, and custom workflows.

Maintenance & Community

Contributions are actively encouraged via a dedicated "Contributing Guide". Users can report bugs through the "Bug reports" link or "Report an Issue" section, and engage in discussions via the "Community" link or "Start a Discussion" section.

Licensing & Compatibility

The project is released under the MIT license, which permits broad use, modification, and distribution, including for commercial purposes, with minimal restrictions.

Limitations & Caveats

Several advanced features, including smart voice notes (summaries, tasks), MCP integration for voice commands, and real-time meeting transcription, are marked as "In Progress" or "Planned," indicating they are not yet available in the stable release. Detailed setup commands and specific hardware recommendations are not explicitly provided in the README.