Zobrazeno 1 - 5
of 5
pro vyhledávání: '"Iman Sarraf Rezaei"'
Publikováno v:
2020 6th Iranian Conference on Signal Processing and Intelligent Systems (ICSPIS).
As a state-of-the-art solution for speaker verification problems, deep neural networks have been usefully employed for extracting speaker embeddings which represent speaker informative features. Objective functions, as the supervisors for the learnin
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::809b184dccc633b86e3fdecbfa7f998f
https://doi.org/10.1109/icspis51611.2020.9349565
https://doi.org/10.1109/icspis51611.2020.9349565
Quality and intelligibility of speech signals are degraded under additive background noise which is a critical problem for hearing aid and cochlear implant users. Motivated to address this problem, we propose a novel speech enhancement approach using
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::d48c49063e2cbb040c0bd03f7eb72fc5
Publikováno v:
7'th International Symposium on Telecommunications (IST'2014).
One of the most important steps in a keyword spotting (KWS) system is a post-processing procedure to compute a confidence measure (CM) for each hypothesized keyword. The CM is commonly estimated by likelihood-based acoustic scores. However durations
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::2bb561751140c8edf5409371a5972de2
https://doi.org/10.1109/istel.2014.7000721
https://doi.org/10.1109/istel.2014.7000721
Autor:
Mohammad Mohsen Goodarzi, Yasser Shekofteh, Jahanshah Kabudian, Iman Sarraf Rezaei, Farshad Almasganj
Publikováno v:
20th Iranian Conference on Electrical Engineering (ICEE2012).
Configuring a whole setup with application of continuous conversational telephony speech recognition in Persian is the goal of this paper. For this propose, two common methods, Gaussian Mixture Model (GMM) and Neural Network (NN) and a proposed hybri
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::c850c4afab18265d2b7f1d3a4969d581
https://doi.org/10.1109/iraniancee.2012.6292541
https://doi.org/10.1109/iraniancee.2012.6292541
Publikováno v:
20th Iranian Conference on Electrical Engineering (ICEE2012).
In traditional keyword spotting (KWS) systems, confidence measure (CM) of each keyword is computed from normalized acoustic likelihoods. In addition to likelihood based scores, some keyword dependent features named predictor features such as duration
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::40ad74dcffb52746e4d7be1e0e9d8e88
https://doi.org/10.1109/iraniancee.2012.6292531
https://doi.org/10.1109/iraniancee.2012.6292531