Estimating vocal tract geometry from acoustic impedance using deep neural network. B T, B., Kapoor, S., & Chen, J. JASA Express Letters, 2(3):034801.
Estimating vocal tract geometry from acoustic impedance using deep neural network [link]Paper  doi  abstract   bibtex   
A data-driven approach using artificial neural networks is proposed to address the classic inverse area function problem, i.e., to determine the vocal tract geometry (modelled as a tube of nonuniform cylindrical cross-sections) from the vocal tract acoustic impedance spectrum. The predicted cylindrical radii and the actual radii were found to have high correlation in the three- and four-cylinder model (Pearson coefficient (q) and Lin concordance coefficient (qc) exceeded 95%); however, for the six-cylinder model, the correlation was low (q around 75% and qc around 69%). Upon standardizing the impedance value, the correlation improved significantly for all cases (q and qc exceeded 90%). VC 2022 Author(s). All article content, except where otherwise noted, is licensed under a Creative Commons Attribution (CC BY) license (http://creativecommons.org/ licenses/by/4.0/).
@article{b_t_estimating_2022,
	title = {Estimating vocal tract geometry from acoustic impedance using deep neural network},
	volume = {2},
	issn = {2691-1191},
	url = {https://asa.scitation.org/doi/10.1121/10.0009599},
	doi = {10.1121/10.0009599},
	abstract = {A data-driven approach using artificial neural networks is proposed to address the classic inverse area function problem, i.e., to determine the vocal tract geometry (modelled as a tube of nonuniform cylindrical cross-sections) from the vocal tract acoustic impedance spectrum. The predicted cylindrical radii and the actual radii were found to have high correlation in the three- and four-cylinder model (Pearson coefficient (q) and Lin concordance coefficient (qc) exceeded 95\%); however, for the six-cylinder model, the correlation was low (q around 75\% and qc around 69\%). Upon standardizing the impedance value, the correlation improved significantly for all cases (q and qc exceeded 90\%). {VC} 2022 Author(s). All article content, except where otherwise noted, is licensed under a Creative Commons Attribution ({CC} {BY}) license (http://creativecommons.org/ licenses/by/4.0/).},
	pages = {034801},
	number = {3},
	journal = {{JASA} Express Letters},
	author = {B T, Balamurali and Kapoor, Saumitra and Chen, Jer-Ming},
	urldate = {2023-02-02},
	date = {2022-03},
	langid = {english},
	file = {B T et al. - 2022 - Estimating vocal tract geometry from acoustic impe.pdf:files/108/B T et al. - 2022 - Estimating vocal tract geometry from acoustic impe.pdf:application/pdf},
}

Downloads: 0