Wiener Filter
Definition
A minimum mean-squared-error filter that computes a frequency-dependent gain function from the signal-to-noise ratio in each frequency band. Less aggressive than spectral subtraction, with fewer musical noise artefacts but similar dependence on a good noise estimate.
- Type
- Minimum mean-squared-error filter
- Input
- Signal-to-noise ratio per frequency band
- Compared to spectral subtraction
- Less aggressive, fewer musical noise artefacts
- Limitation
- Depends on a good noise estimate
Common questions
Why is a Wiener filter often preferred over spectral subtraction for forensic audio enhancement?+
Spectral subtraction can leave behind musical noise, isolated tonal artefacts scattered across the spectrum, that distract from intelligibility, while the Wiener filter's gain function tends to produce smoother, less artefact-prone output at a similar level of noise reduction.
What happens to Wiener filter output when the noise estimate is inaccurate?+
An underestimated noise level leaves residual background noise in the output, while an overestimated level can suppress parts of the desired speech signal, so the quality of the noise estimate directly limits how well the filter performs.
Related terms
- Non-Stationary Noise
- Interference whose spectral content changes over time: a vehicle passing, a crowd noise that rises and falls, or a door slamming. Adaptive...
- PESQ / STOI
- Perceptual Evaluation of Speech Quality and Short-Time Objective Intelligibility: instrumental measures that predict listener intelligibility without needing human subjects. STOI scores between...
- Spectral Subtraction
- An enhancement method that estimates the noise spectrum during a speech-free interval and subtracts it from every subsequent frame. Simple and effective...
- Stationary Noise
- Noise whose spectral characteristics remain approximately constant over the duration of a recording. Tape hiss, HVAC fan noise, and mains hum are...
- SWGDE
- Scientific Working Group for Digital Evidence. A US multi-agency body that publishes consensus best-practice documents for digital forensic disciplines, including image authentication,...