Academic Bibliography

 Search 200 years of publications by Ghent University researchers.
Advanced search
1 file | 416.95 KB
Add to list
  1. Search results
  2. Behavioral analysis of pathological s...

Behavioral analysis of pathological speaker embeddings of patients during oncological treatment of oral cancer

Jenthe Thienpondt (UGent) , Caroline M. Speksnijder and Kris Demuynck (UGent)
(2023) INTERSPEECH 2023. In Interspeech p.3018-3022
Author
Organization
Abstract
In this paper, we analyze the behavior of speaker embeddings of patients during oral cancer treatment. First, we found that pre- and post-treatment speaker embeddings differ significantly, notifying a substantial change in voice characteristics. However, a partial recovery to pre-operative voice traits is observed after 12 months post-operation. Secondly, the same-speaker similarity at distinct treatment stages is similar to healthy speakers, indicating that the embeddings can capture characterizing features of even severely impaired speech. Finally, a speaker verification analysis signifies a stable false positive rate and variable false negative rate when combining speech samples of different treatment stages. This indicates robustness of the embeddings towards other speakers, while still capturing the changing voice characteristics during treatment. To the best of our knowledge, this is the first analysis of speaker embeddings during oral cancer treatment of patients.
Keywords
pathological speaker embeddings, oral cancer treatment, speaker recognition, TONGUE AND/OR FLOOR, SPEECH CONSIDERATIONS, OROPHARYNGEAL CANCER, MALIGNANCIES, SURGERY, MOUTH

Downloads

  • Download
    DS669.pdf
    • full text (Published version)
    • |
    • open access
    • |
    • PDF
    • |
    • 416.95 KB

Citation

Please use this url to cite or link to this publication:

MLA
Thienpondt, Jenthe, et al. “Behavioral Analysis of Pathological Speaker Embeddings of Patients during Oncological Treatment of Oral Cancer.” INTERSPEECH 2023, International Speech Communication Association (ISCA), 2023, pp. 3018–22, doi:10.21437/interspeech.2023-1868.
APA
Thienpondt, J., Speksnijder, C. M., & Demuynck, K. (2023). Behavioral analysis of pathological speaker embeddings of patients during oncological treatment of oral cancer. INTERSPEECH 2023, 3018–3022. https://doi.org/10.21437/interspeech.2023-1868
Chicago author-date
Thienpondt, Jenthe, Caroline M. Speksnijder, and Kris Demuynck. 2023. “Behavioral Analysis of Pathological Speaker Embeddings of Patients during Oncological Treatment of Oral Cancer.” In INTERSPEECH 2023, 3018–22. International Speech Communication Association (ISCA). https://doi.org/10.21437/interspeech.2023-1868.
Chicago author-date (all authors)
Thienpondt, Jenthe, Caroline M. Speksnijder, and Kris Demuynck. 2023. “Behavioral Analysis of Pathological Speaker Embeddings of Patients during Oncological Treatment of Oral Cancer.” In INTERSPEECH 2023, 3018–3022. International Speech Communication Association (ISCA). doi:10.21437/interspeech.2023-1868.
Vancouver
1.
Thienpondt J, Speksnijder CM, Demuynck K. Behavioral analysis of pathological speaker embeddings of patients during oncological treatment of oral cancer. In: INTERSPEECH 2023. International Speech Communication Association (ISCA); 2023. p. 3018–22.
IEEE
[1]
J. Thienpondt, C. M. Speksnijder, and K. Demuynck, “Behavioral analysis of pathological speaker embeddings of patients during oncological treatment of oral cancer,” in INTERSPEECH 2023, Dublin, Ireland, 2023, pp. 3018–3022.
@inproceedings{01HAPD8QHD7J4MTG9Z5KKQD9N5,
  abstract     = {{In this paper, we analyze the behavior of speaker embeddings of patients during oral cancer treatment. First, we found that pre- and post-treatment speaker embeddings differ significantly, notifying a substantial change in voice characteristics. However, a partial recovery to pre-operative voice traits is observed after 12 months post-operation. Secondly, the same-speaker similarity at distinct treatment stages is similar to healthy speakers, indicating that the embeddings can capture characterizing features of even severely impaired speech. Finally, a speaker verification analysis signifies a stable false positive rate and variable false negative rate when combining speech samples of different treatment stages. This indicates robustness of the embeddings towards other speakers, while still capturing the changing voice characteristics during treatment. To the best of our knowledge, this is the first analysis of speaker embeddings during oral cancer treatment of patients.}},
  author       = {{Thienpondt, Jenthe and Speksnijder, Caroline M. and Demuynck, Kris}},
  booktitle    = {{INTERSPEECH 2023}},
  issn         = {{2308-457X}},
  keywords     = {{pathological speaker embeddings,oral cancer treatment,speaker recognition,TONGUE AND/OR FLOOR,SPEECH CONSIDERATIONS,OROPHARYNGEAL CANCER,MALIGNANCIES,SURGERY,MOUTH}},
  language     = {{eng}},
  location     = {{Dublin, Ireland}},
  pages        = {{3018--3022}},
  publisher    = {{International Speech Communication Association (ISCA)}},
  title        = {{Behavioral analysis of pathological speaker embeddings of patients during oncological treatment of oral cancer}},
  url          = {{http://doi.org/10.21437/interspeech.2023-1868}},
  year         = {{2023}},
}
Altmetric
View in Altmetric
Web of Science
Times cited: