Zobrazeno 1 - 8
of 8
pro vyhledávání: '"Mojtaba Hasannezhad"'
Publikováno v:
IEEE Open Journal of Signal Processing, Vol 2, Pp 136-150 (2021)
A natural choice to model strong temporal dynamics of speech is the recurrent neural network (RNN) since it can exploit the sequential information from consecutive acoustic frames and generalizes the model well to unseen speakers. Besides, the convol
Externí odkaz:
https://doaj.org/article/9bcc48c8ccd64219bff6a984db6b61d9
Publikováno v:
Speech Communication. 136:1-13
Most of the current approaches for speech enhancement (SE) using deep neural network (DNN) face a number of limitations: they do not exploit information contained in the phase spectrum while their high computational complexity and memory requirements
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::bf4dec95b309ed86c50175b18e44ffb9
https://doi.org/10.1016/j.specom.2021.10.002
https://doi.org/10.1016/j.specom.2021.10.002
Publikováno v:
IEEE Open Journal of Signal Processing, Vol 2, Pp 136-150 (2021)
A natural choice to model strong temporal dynamics of speech is the recurrent neural network (RNN) since it can exploit the sequential information from consecutive acoustic frames and generalizes the model well to unseen speakers. Besides, the convol
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::5c04033dc85eb64f4aebe2f8a137d4ef
https://doi.org/10.1109/ojsp.2021.3067147
https://doi.org/10.1109/ojsp.2021.3067147
Publikováno v:
ISCAS
Speech exhibits strong dependencies among its samples in both time and frequency domains. In this paper, we propose a low-complexity composite model for speech enhancement (SE) that integrates a convolutional neural network (CNN) and a long short-ter
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::a5bf7a6eadfdf87614a4f37f42e037b4
https://doi.org/10.1109/iscas51556.2021.9401385
https://doi.org/10.1109/iscas51556.2021.9401385
Publikováno v:
MWSCAS
Speech spectrograms exhibit strong contextual dependencies along both time and frequency dimensions. In this paper, a novel composite model integrating a long short-term memory (LSTM) and convolutional neural network (CNN) to exploit temporal and spe
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::62c363261766212c15b40a920be87bdb
https://doi.org/10.1109/mwscas48704.2020.9184645
https://doi.org/10.1109/mwscas48704.2020.9184645
Publikováno v:
7'th International Symposium on Telecommunications (IST'2014).
In this paper, a novel Direct Digital Frequency Synthesizer (DDFS) based on using non-uniform segmentation in sine-weighted Digital-to-Analog Convertor (DAC) is proposed. To generating beyond Nyquist frequency signal, parallel DACs with Return-To-Zer
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::8a262cab6498c3b6a5df88eb7b3645aa
https://doi.org/10.1109/istel.2014.7000689
https://doi.org/10.1109/istel.2014.7000689
Publikováno v:
7'th International Symposium on Telecommunications (IST'2014).
In this paper, a novel Direct Digital Frequency Synthesizer (DDFS) based on using squared-sine-weighted Digital-to-Analog Convertor (DAC) is proposed. To increase the speed, parallel DACs with Return-To-Zero (RTZ) technique are used. In the conventio
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::ec09c8ee9fa52c7d28947d8ecc09c4c6
https://doi.org/10.1109/istel.2014.7000736
https://doi.org/10.1109/istel.2014.7000736
Publikováno v:
Journal of Circuits, Systems and Computers. 25:1650085
This paper presented a low-power Direct Digital Frequency Synthesizer (DDFS) using non-uniform sine-weighted digital-to-analog convertor (DAC). To avoid the need for a sharp filter to generate signals near and beyond the Nyquist frequency, parallel D
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::e22000886a15f443c2bf09a159d8450f
https://doi.org/10.1142/s0218126616500857
https://doi.org/10.1142/s0218126616500857