Výsledky vyhledávání - "Akinobu, Lee"

Context and knowledge aware conversational model and system combination for grounded response generation

Autor: Shugo Kato, Akinobu Lee, Akihide Ozeki, Ryota Tanaka

Publikováno v: Computer Speech & Language. 62:101070

End-to-end neural-based dialogue systems can potentially generate tailored and coherent responses for user inputs. However, most of existing systems produce universal and non-informative responses, and they have not gone beyond chitchat yet. To tackl

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::87f778d9b1e638ab3f91eb170ceaec54
https://doi.org/10.1016/j.csl.2020.101070

Zobrazit plný text záznamu

MMDAgent - A Fully Open-Source Toolkit for Voice Interaction Systems

Autor: Akinobu, Lee, Keiichiro, Oura, Keiichi, Tokuda

Publikováno v: 2013 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2013). :8382-8385

Vancouver, BC, Canada, 26-31 May 2013

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=jairo_______::08c626f7b000391093bdec08657234ad
http://id.nii.ac.jp/1476/00004621/

Zobrazit plný text záznamu

User Generated Dialogue Systems: uDialogue

Autor: Steve Renals, Ichi Takumi, Yoshihiko Nankaku, Keiichi Tokuda, Kei Hashimoto, Shuhei Tsutsumi, Daisuke Yamamoto, Junichi Yamagishi, Akinobu Lee, Keiichiro Oura, Takahiro Uchiya

Publikováno v: Human-Harmonized Information Technology, Volume 2 ISBN: 9784431565338
Human-Harmonized Information Technology (2)

This chapter introduces the idea of user-generated dialogue content and describes our experimental exploration aimed at clarifying the mechanism and conditions that makes it workable in practice. One of the attractive points of a speech interface is

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::ed0a571ca93ef6739a76841e667e8132
https://doi.org/10.1007/978-4-431-56535-2_3

Zobrazit plný text záznamu

Bayesian Context Clustering Using Cross Validation for Speech Recognition

Autor: Kei Hashimoto, Heiga Zen, Yoshihiko Nankaku, Akinobu Lee, Keiichi Tokuda

Publikováno v: IEICE transactions on information and systems. (3):668-678

This paper proposes Bayesian context clustering using cross validation for hidden Markov model (HMM) based speech recognition. The Bayesian approach is a statistical technique for estimating reliable predictive distributions by treating model paramet

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::8e01f7cb3d2f11aaa40eb562b32ec09d
http://id.nii.ac.jp/1476/00005504/

Zobrazit plný text záznamu

Speech recognition based on statistical models including multiple phonetic decision trees

Autor: Akinobu Lee, Keiichi Tokuda, Yoshihiko Nankaku, Kei Hashimoto, Heiga Zen, Sayaka Shiota

Publikováno v: Acoustical Science and Technology. 32:236-243

We propose a speech recognition technique using multiple model structures. In the use of context-dependent models, decision-tree-based context clustering is applied to find an appropriate parameter tying structure. However, context clustering is usua

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::d1578f7643f2991beb2742c55aa9c800
https://doi.org/10.1250/ast.32.236

Zobrazit plný text záznamu

A Covariance-Tying Technique for HMM-Based Speech Synthesis

Autor: Keiichi Tokuda, Yoshihiko Nankaku, Keiichiro Oura, Heiga Zen, Akinobu Lee

Publikováno v: IEICE Transactions on Information and Systems. :595-601

SUMMARY A technique for reducing the footprints of HMM-based speech synthesis systems by tying all covariance matrices of state distributions is described. HMM-based speech synthesis systems usually leave smaller footprints than unit-selection synthe

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::f144027a295b48e068969421411f141e
https://doi.org/10.1587/transinf.e93.d.595

Zobrazit plný text záznamu

Speaker Adaptation Based on Nonlinear Spectral Transform for Speech Recognition

Autor: Keiichi Tokuda, Akinobu Lee, Toyohiro Hayashi, Yoshihiko Nankaku

Publikováno v: INTERSPEECH

This paper proposes a speaker adaptation technique using a nonlinear spectral transform based on GMMs. One of the most popular forms of speaker adaptation is based on linear transforms, e.g., MLLR. Although MLLR uses multiple transforms according to

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::1ac6bb48ed04a5981b569d8ff097bece
http://id.nii.ac.jp/1476/00003402/

Zobrazit plný text záznamu

A Fully Consistent Hidden Semi-Markov Model-Based Speech Recognition System

Autor: Akinobu Lee, Heiga Zen, Yoshihiko Nankaku, Keiichi Tokuda, Keiichiro Oura

Publikováno v: IEICE Transactions on Information and Systems. :2693-2700

In a hidden Markov model (HMM), state duration probabilities decrease exponentially with time, which fails to adequately represent the temporal structure of speech. One of the solutions to this problem is integrating state duration probability distri

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::7b37c19e8b45a19132ff57e44eae2c4b
https://doi.org/10.1093/ietisy/e91-d.11.2693

Zobrazit plný text záznamu

Blind source separation based on a fast-convergence algorithm combining ICA and beamforming

Autor: Toshiya Kawamura, Kiyohiro Shikano, Hiroshi Saruwatari, Akinobu Lee, Tsuyoki Nishikawa

Publikováno v: IEEE Transactions on Audio, Speech and Language Processing. 14:666-678

We propose a new algorithm for blind source separation (BSS), in which independent component analysis (ICA) and beamforming are combined to resolve the slow-convergence problem through optimization in ICA. The proposed method consists of the followin

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::572b470a9424244b0e30a16d8ab5c1eb
https://doi.org/10.1109/tsa.2005.855832

Zobrazit plný text záznamu

Vyhledávací nástroje:

Upřesnit hledání