Zobrazeno 1 - 10 of 60 pro vyhledávání: '"Yongwon Jeong"'
1
Linear prediction-based dereverberation with very deep convolutional neural networks for reverberant speech recognition
Autor: Min Sik Kim, Sunchan Park, Hyung Soon Kim, Yongwon Jeong
Publikováno v: 2018 International Conference on Electronics, Information, and Communication (ICEIC).
Convolutional neural networks (CNNs) have been shown to improve classification tasks such as automatic speech recognition (ASR). Furthermore, the CNN with very deep architecture lowered the word error rate (WER) in reverberant and noisy environments.
Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::234ad8bee58eccad3536a57b990f36cd
https://doi.org/10.23919/elinfocom.2018.8330593
Zobrazit plný text záznamu
2
Basis-Based Speaker Adaptation Using Partitioned HMM Mean Parameters of Training Speaker Models
Autor: Yongwon Jeong
Publikováno v: Journal of Signal Processing Systems. 82:303-310
This paper presents the basis-based speaker adaptation method that includes approaches using principal component analysis (PCA) and two-dimensional PCA (2DPCA). The proposed method partitions the hidden Markov model (HMM) mean vectors of training mod
Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::248492491d8eb1130d3006443c6ba455
https://doi.org/10.1007/s11265-015-0996-2
Zobrazit plný text záznamu
3
Multiresolution CNN for reverberant speech recognition
Autor: Yongwon Jeong, Sunchan Park, Hyung Soon Kim
Publikováno v: O-COCOSDA
The performance of automatic speech recognition (ASR) has been greatly improved by deep neural network (DNN) acoustic models. However, DNN-based systems still perform poorly in reverberant environments. Convolutional neural network (CNN) acoustic mod
Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::4c314520e69b9191571145bcb57f9c89
https://doi.org/10.1109/icsda.2017.8384470
Zobrazit plný text záznamu
4
Joint speaker and environment adaptation using TensorVoice for robust speech recognition
Autor: Yongwon Jeong
Publikováno v: Speech Communication. 58:1-10
We present an adaptation of a hidden Markov model (HMM)-based automatic speech recognition system to the target speaker and noise environment. Given HMMs built from various speakers and noise conditions, we build tensorvoices that capture the interac
Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::61f3399482ff8e0267ef135a143b02d0
https://doi.org/10.1016/j.specom.2013.10.001
Zobrazit plný text záznamu
7
Unified framework for basis-based speaker adaptation based on sample covariance matrix of variable dimension
Autor: Yongwon Jeong
Publikováno v: Speech Communication. 55:340-346
We present a unified framework for basis-based speaker adaptation techniques, which subsumes eigenvoice speaker adaptation using principal component analysis (PCA) and speaker adaptation using two-dimensional PCA (2DPCA). The basic idea is to partiti
Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::b6b0b9928e3d0611422ed786de592735
https://doi.org/10.1016/j.specom.2012.09.002
Zobrazit plný text záznamu
10
Adaptation of Hidden Markov Models Using Model-as-Matrix Representation
Autor: Yongwon Jeong
Publikováno v: IEEE Transactions on Audio, Speech, and Language Processing. 20:2352-2364
In this paper, we describe basis-based speaker adaptation techniques using the matrix representation of training models. Bases are obtained from training models by decomposition techniques for matrix-variate objects: two-dimensional principal compone
Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::0cd8c3867b57cf03667f07c02d2b8b38
https://doi.org/10.1109/tasl.2012.2202649
Zobrazit plný text záznamu

Vyhledávací nástroje:

Upřesnit hledání

Omezení vyhledávání
Plný text Recenzováno Digitální knihovna AV ČR
Zdroje