Perceptual evaluation of speech quality (PESQ)-a new method for speech quality assessment of telephone networks and codecs

By A. Rix, J. Beerends, M. Hollier, A. Hekstra. 2001

3,796 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Electrical and Electronic Engineering, Evaluation & Benchmarks, Speech & Audio, 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 Computer Vision and Pattern Recognition

In Electrical and Electronic Engineering

In Evaluation & Benchmarks

In Speech & Audio

In Signal Processing

Citations in the corpus

Noise Estimation for Generative Diffusion Models (2021, methodology): “Even though our method results in a small decrease in the MCD [18], we demonstrate a large improvement in both PESQ [26] and STOI [33].”
Vocos: Closing the gap between time-domain and Fourier-based neural vocoders for high-quality audio synthesis (2023, methodology): “Our evaluation process also encompasses several other metrics, including the Perceptual Evaluation of Speech Quality (PESQ) (Rix et al., 2001), periodicity error, and the F1 score for voiced/unvoiced classification (V/UV F1), following the methodology proposed by Morrison et al. (2021).”
VQTTS: High-Fidelity Text-to-Speech Synthesis with Self-Supervised VQ Acoustic Feature (2022, methodology): “Also, we can see that the PESQ value of vec2wav is worse than that of the HifiGAN with mel-spectrogram.”
SpeechT5: Unified-Modal Encoder-Decoder Pre-Training for Spoken Language Processing (2021, methodology): “Since lengths of noisy speech utterances are the same as lengths of clean utterances, we measure the test set via speech quality (PESQ) (Rix et al., 2001), extended short-time objective intelligibility (ESTOI) (Jensen and Taal, 2016), and WER to quantify the difﬁculty of noisy speech, as shown in Ta”
ViSQOL v3: An Open Source Production Ready Objective Speech and Audio Metric (2020, background): “When estimating perceived quality, PESQ [1], [2] and POLQA [3], [4] have become standards for speech, and in practice also for general audio, despite being originally designed to target only speech quality.”
Cascaded Cross-Module Residual Learning towards Lightweight End-to-End Speech Coding (2019, methodology): “3 4.22 4.30 CMRL 13.45 16.35 17.18 17.33 3.69 4.21 4.34 4.42 within the target range. 3.2. Objective test We evaluate 500 decoded utterances in terms of SNR and PESQ with wide band extension (P862.2) [34]. Figure 5 (a) shows the effectiveness of CMRL against a system with a single module in terms of”
Conv-TasNet: Surpassing Ideal Time-Frequency Magnitude Masking for Speech Separation (2018, methodology): “n ﬁgure 3 (C). This observation shows that PESQ consistently underestimates MOS for Conv-TasNet separated utterances, which may be due to the dependence of PESQ on the magnitude spectrogram of speech [45] which could produce lower scores for time-domain approaches. TABLE VI PESQ SCORES FOR THE IDEAL”
BigVGAN: A Universal Neural Vocoder with Large-Scale Training (2022, background): “Overall, the publicly available model trained on train-clean-360 scores marginally better in terms of PESQ and periodicity error compared to our checkpoint trained on train-full .”
Dnsmos: A Non-Intrusive Perceptual Objective Speech Quality Metric to Evaluate Noise Suppressors (2020, background): “PESQ and POLQA reasonable PCC and SRCC.”
Full-Capacity Unitary Recurrent Neural Networks (2016, background): “PESQ is the ITU-T standard for telephone voice quality testing [21, 22], and is a popular perceptual quality metric for speech enhancement [23].”
Multitalker Speech Separation With Utterance-Level Permutation Invariant Training of Deep Recurrent Neural Networks (2017, methodology): “The models were evaluated on their potential to improve the Signal-to-Distortion Ratio (SDR) [44] and the Perceptual Evaluation of Speech Quality (PESQ) [49] score, both of which are metrics widely used to evaluate speech enhancement performance for multi-talker speech separation tasks.”
Looking to listen at the cocktail party (2018, neutral): “The reported objective quality scores are PESQ [Rix et al. 2001], STOI [Taal et al.”
SDR – Half-baked or Well Done? (2018, methodology): “While the merits of various objective measures such as PESQ [12], Loizou’s composite measure [13], PEMO-Q [14], PEASS [15], or STOI [16], could be debated and compared, we are concerned here with an issue with the way the widely relied upon BSS eval tool-box [17] has been used.”
ProDiff: Progressive Fast Diffusion Model for High-Quality Text-to-Speech (2022, methodology): “For objective evaluation, ProDiff also demonstrates the outperformed performance in MCD, PESQ, and STOI, superior to all baseline models.”
UnivNet: A Neural Vocoder with Multi-Resolution Spectrogram Discriminators for High-Fidelity Waveform Generation (2021, methodology): “[29] A. W. Rix, J. G. Beerends, M. P. Hollier, and A. P. Hekstra, “Perceptual Evaluation of Speech Quality (PESQ)-”
Bilateral Denoising Diffusion Models (2021, positive): “In addition, to measure the noisiness and the distortion of the generated speech relative to the reference speech, we adopted two commonly used metrics in speech enhancement – the perceptual evaluation of speech quality (PESQ) [44] and the short-time objective intelligibility (STOI) measure [45].”