CoChat

Using SincNet for Learning Pathological Voice Disorders

Open access

The DOI matches its Crossref record; 33 of the 33 references it cites were found in a registry or an index.

Abstract

Deep learning techniques such as convolutional neural networks (CNN) have been successfully applied to identify pathological voices. However, the major disadvantage of using these advanced models is the lack of interpretability in explaining the predicted outcomes. This drawback further introduces a bottleneck for promoting the classification or detection of voice-disorder systems, especially in this pandemic period. In this paper, we proposed using a series of learnable sinc functions to replace the very first layer of a commonly used CNN to develop an explainable SincNet system for classifying or detecting pathological voices. The applied sinc filters, a front-end signal processor in SincNet, are critical for constructing the meaningful layer and are directly used to extract the acoustic features for following networks to generate high-level voice information. We conducted our tests on three different Far Eastern Memorial Hospital voice datasets. From our evaluations, the proposed approach achieves the highest 7%–accuracy and 9%–sensitivity improvements from conventional methods and thus demonstrates superior performance in predicting input pathological waveforms of the SincNet system. More importantly, we intended to give possible explanations between the system output and the first-layer extracted speech features based on our evaluated results.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.3390/s22176634

Cites

33 of 33 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1 Waveform and Spectrogram of pathological voices.
Figure 1 Waveform and Spectrogram of pathological voices.CC BY · PubMed Central
Figure 2 Architecture of SincNet.
Figure 2 Architecture of SincNet.CC BY · PubMed Central
Figure 3 The confusion matrix is used to show detailed detection performance of (a) CNN(1D) and (b) SincNet on the 2018 FEMH Challenge dataset.
Figure 3 The confusion matrix is used to show detailed detection performance of (a) CNN(1D) and (b) SincNet on the 2018 FEMH Challenge dataset.CC BY · PubMed Central
Figure 4 Scatter plots of (a) CNN(1D) and (b) SincNet show the t-SNE performance for detecting normal and abnormal voice on the 2018 FEMH Challenge dataset.
Figure 4 Scatter plots of (a) CNN(1D) and (b) SincNet show the t-SNE performance for detecting normal and abnormal voice on the 2018 FEMH Challenge dataset.CC BY · PubMed Central
Figure 5 The confusion matrix is used to show detailed detection performance of (a) CNN(1D) and (b) SincNet on the FEMH dataset.
Figure 5 The confusion matrix is used to show detailed detection performance of (a) CNN(1D) and (b) SincNet on the FEMH dataset.CC BY · PubMed Central
Figure 6 Scatter plots of (a) CNN(1D) and (b) SincNet show the t-SNE performance for detecting normal and abnormal voice on the FEMH dataset.
Figure 6 Scatter plots of (a) CNN(1D) and (b) SincNet show the t-SNE performance for detecting normal and abnormal voice on the FEMH dataset.CC BY · PubMed Central
Figure 7 Confusion matrices for (a) CNN(1D) and (b) SincNet are used to show detailed classification performance among Neo, Pho, and VP on the 2018 FEMH Challenge dataset.
Figure 7 Confusion matrices for (a) CNN(1D) and (b) SincNet are used to show detailed classification performance among Neo, Pho, and VP on the 2018 FEMH Challenge dataset.CC BY · PubMed Central
Figure 8 Scatter plots of (a) CNN(1D) and (b) SincNet show the t-SNE performance for classifying Neo, Pho, and VP on the 2018 FEMH Challenge dataset.
Figure 8 Scatter plots of (a) CNN(1D) and (b) SincNet show the t-SNE performance for classifying Neo, Pho, and VP on the 2018 FEMH Challenge dataset.CC BY · PubMed Central
Figure 9 Confusion matrices for (a) CNN(1D) and (b) SincNet are used to show detailed classification performance among Neo, Pho, and VP on the 2019 FEMH Challenge dataset.
Figure 9 Confusion matrices for (a) CNN(1D) and (b) SincNet are used to show detailed classification performance among Neo, Pho, and VP on the 2019 FEMH Challenge dataset.CC BY · PubMed Central
Figure 10 Scatter plots of (a) CNN(1D) and (b) SincNet show the t-SNE performance for classifying Neo, Pho, and VP on the 2019 FEMH Challenge dataset.
Figure 10 Scatter plots of (a) CNN(1D) and (b) SincNet show the t-SNE performance for classifying Neo, Pho, and VP on the 2019 FEMH Challenge dataset.CC BY · PubMed Central
Figure 11 Confusion matrices for (a) CNN(1D) and (b) SincNet are used to show detailed classification performance among Neo, Pho, and VP on the FEMH dataset.
Figure 11 Confusion matrices for (a) CNN(1D) and (b) SincNet are used to show detailed classification performance among Neo, Pho, and VP on the FEMH dataset.CC BY · PubMed Central
Figure 12 Scatter plots of (a) CNN(1D) and (b) SincNet show the t-SNE performance for classifying Neo, Pho, and VP on the FEMH dataset.
Figure 12 Scatter plots of (a) CNN(1D) and (b) SincNet show the t-SNE performance for classifying Neo, Pho, and VP on the FEMH dataset.CC BY · PubMed Central
Figure 13 Confusion matrices for (a) CNN(1D) and (b) SincNet are used to show detailed classification performance among FD, Neo, Pho, and VP on the 2019 FEMH Challenge dataset.
Figure 13 Confusion matrices for (a) CNN(1D) and (b) SincNet are used to show detailed classification performance among FD, Neo, Pho, and VP on the 2019 FEMH Challenge dataset.CC BY · PubMed Central
Figure 14 Scatter plots of (a) CNN(1D) and (b) SincNet show the t-SNE performance for classifying FD, Neo, Pho, and VP using 2019 FEMH Challenge dataset.
Figure 14 Scatter plots of (a) CNN(1D) and (b) SincNet show the t-SNE performance for classifying FD, Neo, Pho, and VP using 2019 FEMH Challenge dataset.CC BY · PubMed Central
Figure 15 Confusion matrices for (a) CNN(1D) and (b) SincNet are used to show detailed classification performance among FD, Neo, Pho, and VP on the FEMH dataset.
Figure 15 Confusion matrices for (a) CNN(1D) and (b) SincNet are used to show detailed classification performance among FD, Neo, Pho, and VP on the FEMH dataset.CC BY · PubMed Central
Figure 16 Scatter plots of (a) CNN(1D) and (b) SincNet show the t-SNE performance for classifying FD, Neo, Pho, and VP using FEMH dataset.
Figure 16 Scatter plots of (a) CNN(1D) and (b) SincNet show the t-SNE performance for classifying FD, Neo, Pho, and VP using FEMH dataset.CC BY · PubMed Central
Figure 17 During training SincNet and CNN(1D) models in (a) 2018 FEMH Challenge and (b) FEMH datasets, lose curves were depicted in this figure to demonstrate the training efficiency.
Figure 17 During training SincNet and CNN(1D) models in (a) 2018 FEMH Challenge and (b) FEMH datasets, lose curves were depicted in this figure to demonstrate the training efficiency.CC BY · PubMed Central
Figure 18 Two channels selected from the input layer of the optimized CNN(1D) and SincNet models were used to demonstrate the filter properties. The upper row shows filters in the time domain, while the bottom row depicts the magnitude components of filters in the frequency domain. (a) CNN Flters35,72; (b) Sinc Flters35,72.
Figure 18 Two channels selected from the input layer of the optimized CNN(1D) and SincNet models were used to demonstrate the filter properties. The upper row shows filters in the time domain, while the bottom row depicts the magnitude components of filters in the frequency domain. (a) CNN Flters35,72; (b) Sinc Flters35,72.CC BY · PubMed Central
Figure 19 The power spectral density of “CNN Filters” and “Sinc Filters” processed pathological voices. Two utterances individually selected from (a) 2018 FEMH Challenge and (b) FEMH datasets were involved in this test.
Figure 19 The power spectral density of “CNN Filters” and “Sinc Filters” processed pathological voices. Two utterances individually selected from (a) 2018 FEMH Challenge and (b) FEMH datasets were involved in this test.CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 33 references

Research works printed but not linked: 6

  1. Hsu, Y.T., Zhu, Z., Wang, C.T., Fang, S.H., Rudzicz, F., and Tsao, Y. (2018). Robustness against the channel effect in pathological voice detection. arXiv.
  2. IEEE (2022, August 10). FEMH Challenge 2018. Available online: https://femh-challenge2018.weebly.com/.
  3. Doran, D., Schulz, S., and Besold, T.R. (2017). What Does Explainable AI Really Mean? A New Conceptualization of Perspectives. arXiv.
  4. Hoffman, R.R., Mueller, S.T., Klein, G., and Litman, J. (2018). Metrics for Explainable AI: Challenges and Prospects. arXiv.
  5. Holzinger, A., Biemann, C., Pattichis, C.S., and Kell, D.B. (2017). What do we need to build explainable AI systems for the medical domain?. arXiv.
  6. Ravanelli, M., and Bengio, Y. (2018). Interpretable Convolutional Filters with SincNet. arXiv.

Cited by

See all 40

Metadata

Authors
Hung, C.-H., Wang, S.-S., Wang, C.-T., & Fang, S.-H.
Affiliations
  • Yuan Ze University (TW)
  • Far Eastern Memorial Hospital (TW)
Corresponding author
Fang, Shih-Hau
Venue
Sensors
Publisher
MDPI AG
Published
2022-09-02
Volume
22
Issue
17
Pages
6634
ISSN
1424-8220 (ISSN-L)
OpenAlex
W4295079846
Cited by
39
References
33
Citations by year
2026: 10 · 2025: 12 · 2024: 10 · 2023: 6 · 2022: 1
FWCI
4.49
Type
journal article
Language
en
Open access
Locations
  • Sensors (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
National Science and Technology Council (106-2314-B-418-003, 111-2622-E-155-002, 111-2221-E-155-020)
Topics
  • Health Sciences › Health Professions › Speech and Hearing › Voice and Speech Disorders
  • Physical Sciences › Computer Science › Artificial Intelligence › Speech Recognition and Synthesis
  • Physical Sciences › Computer Science › Signal Processing › Speech and Audio Processing
Keywords
SincNet, pathological voice detection, voice disorders, explainable deep learning, convolutional neural networks, audio feature extraction, speech processing
MeSH terms
Acoustics, Humans, Voice, Voice Disorders, Neural Networks, Computer

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Using SincNet for Learning Pathological Voice Disorders. https://app.cochat.ai/papers/10.3390/s22176634
The paper
Hung, C.-H., Wang, S.-S., Wang, C.-T., & Fang, S.-H. (2022). Using SincNet for Learning Pathological Voice Disorders. Sensors. https://doi.org/10.3390/s22176634

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.