Module preprocessing 

Audio preprocessing pipeline for improving VAD accuracy.

This module provides configurable preprocessing stages that clean audio before voice activity detection. Each stage is optional and can be enabled/disabled via PreprocessorConfig.

Example

use wavekat_vad::preprocessing::{Preprocessor, PreprocessorConfig};

let config = PreprocessorConfig {
    high_pass_hz: Some(80.0),
    ..Default::default()
};

let mut preprocessor = Preprocessor::new(&config, 16000);
let samples: Vec<i16> = vec![0; 320]; // 20ms at 16kHz
let cleaned = preprocessor.process(&samples);

Structs

AudioResampler

Resampler for converting between sample rates.

BiquadFilter

Second-order (biquad) IIR filter.

Denoiser

RNNoise-based noise suppressor.

Normalizer

Audio normalizer that adjusts amplitude to a target dBFS level.

Preprocessor

Audio preprocessor that applies configured processing stages.

PreprocessorConfig

Configuration for the audio preprocessor.

Constants

DENOISE_SAMPLE_RATE

Internal sample rate required by RNNoise (48 kHz).