Expand description
§ferraus-denoise
Noise reduction: learn a noise fingerprint from a region of the recording, then remove it from the whole file with a soft spectral gate; or track the noise floor adaptively; or run a speech model (DeepFilterNet 3 through ONNX Runtime by default, RNNoise built in).
Commands: denoise, denoise profile, denoise profile merge.
§DeepFilterNet 3
--model dfn3 (the default, ADR 0006) runs the upstream ONNX export (denoise/dfn3:
encoder, ERB decoder, DF decoder) on ONNX Runtime through ferraus-ml, with the STFT,
feature normalization, ERB gains and deep filtering implemented in dfn3 after the
reference libDF/enhance.py (output matches the reference to about 1e-6). The model
comes from the model cache under the download policy of ADR 0004: when it is missing and
downloads are off, the command fails with exit 5 and ferraus models pull denoise/dfn3.
It never falls back to another model silently; --model rnnoise is the built-in,
download-free alternative. Builds without the dfn3 feature default to RNNoise.
Modules§
- dfn3
- DeepFilterNet 3 (Schröter et al., 2023) on ONNX Runtime.
- engine
- The spectral gating engine shared by profile and adaptive modes. It follows the PRD’s seven steps:
- metrics
- Quality metrics for denoising and a naive reference gate.
- ml
- Model-based denoising: RNNoise (pure-Rust
nnnoiseless, weights embedded) and the model slots for DeepFilterNet 3, music-aware denoise, and de-reverb checkpoints. - profile
- Noise profiles: learning, automatic noise-region detection, the
.noiseprofJSON format, and merging. - stft
- STFT framing shared by the profile learner, the gate engine, and the metrics.
Functions§
- ops
- Ops this crate contributes to the command registry.