Preferencje help
Widoczny [Schowaj] Abstrakt
Liczba wyników

Znaleziono wyników: 3

Liczba wyników na stronie
first rewind previous Strona / 1 next fast forward last
Wyniki wyszukiwania
Wyszukiwano:
w słowach kluczowych:  linear filter
help Sortuj według:

help Ogranicz wyniki do:
first rewind previous Strona / 1 next fast forward last
EN
This work aims to further compensate for the weaknesses of feature sparsity and insufficient discriminative acoustic features in existing short-duration speaker recognition. To address this issue, we propose the Bark-scaled Gauss and the linear filter bank superposition cepstral coefficients (BGLCC), and the multidimensional central difference (MDCD) acoustic feature extracted method. The Bark-scaled Gauss filter bank focuses on low-frequency information, while linear filtering is uniformly distributed, therefore, the filter superposition can obtain more discriminative and richer acoustic features of short-duration audio signals. In addition, the multi-dimensional central difference method captures better dynamics features of speakers for improving the performance of short utterance speaker verification. Extensive experiments are conducted on short-duration text-independent speaker verification datasets generated from the VoxCeleb, SITW, and NIST SRE corpora, respectively, which contain speech samples of diverse lengths, and different scenarios. The results demonstrate that the proposed method outperforms the existing acoustic feature extraction approach by at least 10% in the test set. The ablation experiments further illustrate that our proposed approaches can achieve substantial improvement over prior methods.
EN
In this paper analysis of impulsive response and stability conditions of first order linear parametric filters has been carried out. The parameters of these filters have been described by some non-periodic functions. The affiliation of models of LTV (linear time-varying) filters to class of frozen systems (system with slowly varying coefficients) has been shown. In this article the BIBO (bounded input bounded output) stability of LTV filters for strictly positive parametric function has been proved using the generalized eigenvalue method.
3
Content available remote Filtry dla próbek skończonych
PL
Przedstawiono sposoby aproksymacji filtrów linowych o funkcjach odpowiedzi nieskończonej długości. W praktyce, w przypadku stosowania takich filtrów nie dysponujemy odpowiednią liczbą danych - stąd potrzeba odpowiedniego ich „obcięcia" uwzględniającego własności spektralne wejściowej próbki. Zaprezentowano twierdzenia dotyczące filtracji procesów stacjonarnych ARMA oraz procesów ARIMA.
EN
This paper descńbes how linear filter can be optimally implemented for finite time series. Applying linear filter requires a dataset of infinite length, so in practice some sort of approximation is needed. It is shown that the truncated filter needs to be adjusted according to the spectral properties of the ftltered signal.
first rewind previous Strona / 1 next fast forward last
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.