An Algorithm for Predicting the Intelligibility of Speech Masked by Modulated Noise Maskers

By Jesper Højvang Jensen, Cees H. Taal. 2016

729 citations recorded by the source corpus.

Subfields: Biomedical Engineering, Cognitive Neuroscience, 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 Biomedical Engineering

In Cognitive Neuroscience

In Signal Processing

Citations in the corpus

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”
Supervised Speech Separation Based on Deep Learning: An Overview (2017, neutral): “In recent years, the most commonly 380 used intelligibility metric is STOI (short-time objective intelli- 381 gibility), which measures the correlation between the short-time 382 temporal envelopes of a reference (clean) utterance and a sepa- 383 rated utterance [89], [158].”
An Overview of Deep-Learning-Based Audio-Visual Speech Enhancement and Separation (2020, methodology): “An extension of STOI, ESTOI, was later proposed [124] to provide a more accurate prediction of speech intelligibility in presence of highly modulated noise sources.”
Learning Individual Speaking Styles for Accurate Lip to Speech Synthesis (2020, methodology): “We compare our models with the previous lip to speech works using three standard speech quality metrics: Short-Time Objective Intelligibility (STOI) [32] and Extended ShortTime Objective Intelligibility (ESTOI) [17] for estimating the intelligibility and Perceptual Evaluation of Speech Quality (PESQ”
EasyCom: An Augmented Reality Dataset to Support Algorithms for Easy Communication in Noisy Environments (2021, positive): “SIIB has been show to have higher correlation to intelligibility than STOI, ESTOI and HASPI version 1 [30] across many different datasets [31].”
Divide and Conquer: A Deep CASA Approach to Talker-Independent Monaural Speaker Separation (2019, background): “In addition, deep CASA yields the best results in terms of ∆SI-SNR, PESQ and ESTOI.”
SpecGrad: Diffusion Probabilistic Model based Neural Vocoder with Adaptive Noise Spectral Shaping (2022, methodology): “The differences of the scores between the proposed and conventional methods were smaller compared to those of the previous experiments, which should be because the frontend limited the maximum possible quality of the output signals (note that both ESTOI and WARP-Q measure the quality relative to the”
Generative Pre-training for Speech with Flow Matching (2023, methodology): “We report Perceptual Evaluation of Speech Quality (PESQ; Rix et al., 2001), Extended Short-Time Objective Intelligibility (ESTOI; Jensen & Taal, 2016), and Composite Ob-jective Speech Quality and Overall Quality (CSIG/COVL;Hu & Loizou, 2007).”
Speech Enhancement and Dereverberation With Diffusion-Based Generative Models (2022, background): “c) ESTOI: The Extended Short-Time Objective Intelligibility (ESTOI) is an instrumental measure for predicting intelligibility of speech subjected to various kinds of degradation [56].”
Continuous synthesis of artificial speech sounds from human cortical surface recordings during silent speech production (2023, methodology): “Our choice of similarity metrics was motivated by the fact that DTW was preferred over audio spectrogram correlation [30, 40] in a comparable study [20].”