EUVIMEDEUVIMED entdecken
Crossref

On an ARX-LF speech analysis based on TV-CAR analysis

Keiichi Funaki

The Journal of the Acoustical Society of America · 01.10.2025 · 158 (4_Supplement) · A468-A469

DOI: 10.1121/10.0041559 ↗

Kurz gefasst

Fachartikel in der Zeitschrift The Journal of the Acoustical Society of America, erschienen am 01.10.2025. Verfasst von Keiichi Funaki.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

ARX-LF analysis, which simultaneously estimates the glottal excitation model, Liljencrants-Fant model (LF-model), and AR vocal tract filter coefficients from the speech signal, assuming that the speech production model is the source filter model of the glottal excitation and vocal tract filter, has been proposed. This method uses the SEDREAMS method to estimate the accurate GCI (glottal closure instance point), the LF model parameters estimated by the DyProg method from glottal excitation estimated by the IAIF method using Dynamic Programming in several pitch cycles, as preliminary estimates, and the LF model parameters are estimated to identify the ARX model in the neighborhood. On the other hand, we use the TV-CAR analysis for analytic signals, which is a complex-valued signal whose real part is the observed signal and the imaginary part is the Hilbert transform signal of the real one. We have proposed the SEDREAMS method, the IAIF method, and the DyProg method based on the TV-CAR analysis as applications to analytical signals. This presentation shows the construction of the ARX-LF analysis based on these methods.

Zitieren

APA

Funaki, K. (2025). On an ARX-LF speech analysis based on TV-CAR analysis. The Journal of the Acoustical Society of America, 158(4_Supplement), A468-A469. https://doi.org/10.1121/10.0041559

Vancouver

Funaki K. On an ARX-LF speech analysis based on TV-CAR analysis. The Journal of the Acoustical Society of America. 2025;158(4_Supplement):A468-A469. doi:10.1121/10.0041559

BibTeX
@article{Funaki2025,
  author = {Funaki, Keiichi},
  title = {On an ARX-LF speech analysis based on TV-CAR analysis},
  journal = {The Journal of the Acoustical Society of America},
  year = {2025},
  volume = {158},
  number = {4_Supplement},
  pages = {A468--A469},
  doi = {10.1121/10.0041559}
}
Kostenlos starten →