Masking (illustration)

Quality (philosophy)

Function (biology)

Philosophy

Neuroscience

Biology

Bandwidth (computing)

Statistics

Auditory masking

Distortion (music)

Economics

Psychoacoustics

Amplifier

Computer science

Visual arts

Mean opinion score

Evolutionary biology

Computer network

PESQ

Speech enhancement

Psychology

Pattern recognition (psychology)

Mathematics

Epistemology

Operations management

Artificial intelligence

Deep learning

Metric (unit)

Artificial neural network

Noise reduction

Perception

Mean squared error

Speech recognition

IEEE Signal Processing Letters

This letter proposes a perceptual metric for speech quality evaluation, which is suitable, as a loss function, for training deep learning methods. This metric, derived from the perceptual evaluation of the speech quality algorithm, is computed in a per-frame basis and from the power spectra of the reference and processed speech signal. Thus, two disturbance terms, which account for distortion once auditory masking and threshold effects are factored in, amend the mean square error (MSE) loss function by introducing perceptual criteria based on human psychoacoustics. The proposed loss function is evaluated for noisy speech enhancement with deep neural networks. Experimental results show that our metric achieves significant gains in speech quality (evaluated using an objective metric and a listening test) when compared to using MSE or other perceptual-based loss functions from the literature.

A Deep Learning Loss Function Based on the Perceptual Evaluation of the Speech Quality