System audio to text¶
Record system audio and automatically transcribe to text using ✨AI✨.
Overview¶
sys2txt is a command-line tool that records your system audio (via PulseAudio/PipeWire monitor sources) with ffmpeg and transcribes it locally using Whisper. It supports both:
- On-demand: Record until you stop, then transcribe once
- Live-ish: Segment the recording every N seconds and transcribe each segment as it's created (prints continuously)
You can use any of three transcription engines:
faster-whisper- Default, best for CPU and NVIDIA GPUsopenai-whisper- Reference Python implementationwhisper.cpp- C++ implementation with Vulkan GPU support for AMD GPUs
The tool auto-selects the first engine that is installed, preferring faster-whisper for its speed,
then openai-whisper, then whisper.cpp. With none of them installed it says so, rather than
failing part-way through one of them.
See Installation to get started, or jump straight to Quick start.