Skip to main content

Welcome to RFWhisper

Real-time AI denoising for ham radio. No cloud. No compromises. Just clean QSOs from the noisy RF aether.

RFWhisper is a fully local, open-source, real-time ML-powered noise-reduction tool built specifically for amateur radio. It wraps a ham-tuned deep learning denoiser (DeepFilterNet3 primary, RNNoise fallback) inside a GNU Radio + SoapySDR + ONNX Runtime pipeline so you can pull weak signals out of the modern RFI soup — without streaming your audio to a cloud service you don't trust.

Everything runs on your machine. Your shack. Your GPU. Your rules. 73

What you get​

  • Offline WAV denoising with before/after reports (SNR gain, spectrograms, decoder pass-through counts).
  • Real-time audio → virtual cable → WSJT-X, fldigi, JS8Call, SDR#, Quisk.
  • GNU Radio flowgraphs for RTL-SDR, Airspy, Pluto, HackRF, SDRplay (v0.2).
  • Mode-aware profiles: SSB, CW, FT8/FT4, RTTY, VHF FM (v0.3).
  • Before/after spectrogram UI with live SNR/latency/CPU telemetry (v0.4).
  • Fine-tuning tools to adapt the model to your noise environment (v0.5).
  • GPLv3 — inspect, modify, fork, redistribute. Forever.

What you don't get​

  • A cloud dependency. Ever. Core stays 100% local.
  • Silent telemetry. Anything optional is opt-in and disclosed.
  • A denoiser that damages your signals. Every model candidate ships only after passing hard CI gates for CW-transient preservation and FT8 decode counts.

Where to next​

  • New here? → Why RFWhisper explains the real-world noise problem we're solving.
  • Ready to try it? → Quick Start gets you denoising a WAV in under 5 minutes.
  • Want to test it properly? → v0.1 Test Guide walks through the full acceptance suite.
  • Contributing? → Architecture and AGENTS.md are your entry points.
For hams in a hurry

If you know what a virtual audio cable is, install RFWhisper, route your rig audio through it, and point WSJT-X at the output. You'll probably see the value in under 10 minutes. We'll still be here when you want the deep dive.