Intrusive and Non-Intrusive Perceptual Speech Quality Assessment Using a Convolutional Neural Network

By Hannes Gamper, Chandan K. A. Reddy, Ross Cutler, Ivan J. Tashev, Johannes Gehrke. 2019

61 citations recorded by the source corpus.

Speech quality, as perceived by humans, is an important performance metric for telephony and voice services. It is typically measured through subjective listening tests, which can be tedious and expensive. Algorithms such as PESQ and POLQA serve as a computational proxy for subjective listening tests. Here we propose using a convolutional neural network to predict the perceived quality of speech with noise, reverberation, and distortions, both intrusively and non-intrusively, i.e., with and without a clean reference signal. The network model is trained and evaluated on a corpus of about ten thousand utterances labeled by human listeners to derive a Mean Opinion Score (MOS) for each utterance. It is shown to provide more accurate MOS estimates than existing speech quality metrics, including PESQ and POLQA. The proposed method reduces the root mean squared error from 0.48 to 0.35 MOS points and increases the Pearson correlation from 0.78 to 0.89 compared to the state-of-the-art POLQA algorithm.

Subfields: Cognitive Neuroscience, Computational Mechanics, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Cognitive Neuroscience

1. Vishak Gopal
2. Ross Cutler
3. Chandan K. A. Reddy
4. Jan Skoglund
5. Hannes Gamper
6. Sebastian Braun
7. Robert Aichner
8. Harishchandra Dubey

In Computational Mechanics

1. Vishak Gopal
2. Ross Cutler
3. Chandan K. A. Reddy
4. Andrew Hines
5. Robert Aichner
6. Hannes Gamper
7. Harishchandra Dubey
8. Sebastian Braun

In Signal Processing

1. Ross Cutler
2. Vishak Gopal
3. Chandan K. A. Reddy
4. Robert Aichner
5. Hannes Gamper
6. Sebastian Braun
7. Jan Skoglund
8. Harishchandra Dubey

Citations in the corpus

• ViSQOL v3: An Open Source Production Ready Objective Speech and Audio Metric (2020, methodology): “Since then, deep neural network models have emerged and been applied to speech quality models [14], [15].”
• Dnsmos: A Non-Intrusive Perceptual Objective Speech Quality Metric to Evaluate Noise Suppressors (2020, methodology): “Pearson Correlation Coefficient (PCC) or MSE between the predictions of the developed objective metric and the ground truth human ratings is commonly used to measure the accuracy of the model [8, 9].”
• Human Listening and Live Captioning: Multi-Task Training for Speech Enhancement (2021, methodology): “We evaluated our models using PESQ [34], STOI [35], SDR [36], and pMOS [37] metrics, where pMOS is a neuralnetwork-based non-intrusive MOS estimator that shows high correlations with the human MOS ratings without requiring reference signals.”
• ICASSP 2023 Deep Noise Suppression Challenge (2023, background): “Though, this was not a requirement for this challenge.”