hns is a speech-to-text command-line tool that transcribes microphone input directly to the system clipboard. It operates entirely offline on consumer-grade hardware without subscriptions or data transmission to external services.
The tool addresses the need for local voice input in terminal-based workflows. It records speech from a microphone and converts it to text placed on the clipboard, enabling seamless pasting into other applications. Integration examples include piping output to tools such as Claude Code, Ollama, and other command-line interfaces for LLM-driven tasks. Specific use cases named are voice programming, where detailed prompts are provided to coding agents like Claude Code, Codex CLI, or Cursor CLI, and voice journaling to capture thoughts quickly.
As a terminal-native utility, hns follows the Unix philosophy of performing one task well. It runs via a simple installation command using uv tool install hns and is positioned for developers who combine it with existing CLI tools and scripts. Data remains on the local machine, avoiding cloud processing, model training, or third-party sharing.
The project maintains documentation and a GitHub repository for further details. It is presented as suitable for scenarios requiring fast, private speech capture without recurring costs.
In the CLI tools & terminal space, hns takes a focused approach. It focuses on transcribing spoken thoughts and instructions into text quickly and privately without relying on cloud services or paying subscriptions. It is built as a consumer product for developers. It is available for free. It runs on the command line.
It is developed by primaprashant, and it first shipped in 2025. Development happens publicly on GitHub with 113 stars and 16 commits in the last 90 days. Among its 5 catalogued features are Offline Transcription, voice to Clipboard, and Microphone Input.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Same category — not a similarity match