asr2clip

A real-time speech-to-text clipboard tool.

asr2clip recognizes speech in real-time, converts it to text, and automatically copies the result to your system clipboard. It leverages ASR API services for speech recognition and provides flexible recording modes for different use cases.

Quick Start¶

pip install asr2clip       # Install the package
asr2clip --edit            # Create/edit config file
asr2clip --test            # Test your configuration
asr2clip                   # Start recording and transcribing

Key Features¶


Real-time Transcription	Record and transcribe speech with a single command
Clipboard Integration	Automatically copy results to clipboard (no external tools needed)
Voice Activity Detection	Multi-feature VAD with adaptive threshold for auto-transcription
Continuous Mode	Long recording for meetings and lectures
Multiple Backends	OpenAI Whisper, SiliconFlow, Xinference, and more
Local ASR Server	Optional offline transcription with sherpa-onnx
File Transcription	Transcribe existing audio files
Minimal Dependencies	Core functionality with vendored YAML and HTTP modules
Cross-platform	Works on Linux, macOS, and Windows

Use Cases¶

Meeting transcription — Record meetings with continuous mode and VAD, automatically transcribing speech segments to a file with timestamps.

Quick voice notes — Press a key, speak, and have the text instantly in your clipboard ready to paste anywhere.

Lecture notes — Capture lecture audio with automatic segmentation and transcription.

Audio file processing — Transcribe existing audio files from the command line.

Documentation¶

Installation — System dependencies and installation methods
Quick Start — Get up and running in minutes
Configuration — Config file format and options
Usage Guide — Recording modes, VAD, and advanced features
API Reference — Module documentation
Changelog — Version history

License¶

GNU Affero General Public License v3.0