PESQ / STOI
Definition
Perceptual Evaluation of Speech Quality and Short-Time Objective Intelligibility: instrumental measures that predict listener intelligibility without needing human subjects. STOI scores between 0 and 1 correlate well with the fraction of words correctly understood in noise.
- PESQ measures
- Perceived overall speech quality
- STOI measures
- Speech intelligibility
- STOI score range
- 0 to 1
- Advantage
- No human listening panel required
Common questions
What is the practical difference between PESQ and STOI?+
PESQ predicts how good a listener would judge the audio to sound, factoring in distortion and naturalness, while STOI specifically predicts how much of the spoken content a listener would correctly understand, which matters more for enhancement aimed at intelligibility rather than pleasantness.
How well does a STOI score correlate with real word recognition rates?+
STOI was designed and validated to track the proportion of words correctly identified by listeners in noisy conditions closely enough that it is used as a proxy for intelligibility testing without running new listener panels for every recording.
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...
- 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,...
- Wiener Filter
- 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...