Papers - SUGIURA, Komei

Service-Controlled Networking: アプリケーション構 やユーザ要求に連動したネットワーク制御技術

木全崇， 杉浦孔明， 董冕雄， 是津耕司

 2014

RPD-HMM に基づく逐次動作生成による物体操作の模倣学習

杉浦孔明， 是津耕司

人工知能学会全国大会論文集 第 28 回全国大会 (2014), 1I5OS09b5-1I5OS09b5 (The Japanese Society for Artificial Intelligence)  2014 ( 0 ) 1I5OS09b5 - 1I5OS09b5 2014

　View Summary

<p>サービスロボットにとって物体操作は必要不可欠であるが，時々刻々と変化する状況に応じた動作を事前にプログラムするコストは大きい．これまで我々はRPD-HMMを用いた模倣学習手法を提案してきたが，逐次的に動作が生成できないという問題があった．本論文ではRPD-HMMから生成された最尤軌道を基準として逐次的に動作生成を行なう手法を提案する．</p>

• Access to Document (DOI)

Quality-of-experience (QoE) in emerging mobile social networks

M Dong, T Kimata, K Sugiura, K Zettsu

IEICE TRANSACTIONS on Information and Systems 97 (10), 2606-2612 (一般社団法人 電子情報通信学会)  97 ( 10 ) 2606 - 2612 2014

ISSN  0916-8532

　View Summary

Mobile social networks (MSN) provides diverse services to meet the needs of mobile users, i.e., discovering new friends, and sharing their pictures, videos and other information among their common interest friends. On the other hand, Quality-of-Experience (QoE) is a new concept related to but differs from Quality-of-Service (QoS) perception. QoE is a subjective measure of a customer's experiences with a service focuses on the entire service experience, and is a more holistic evaluation. So far, QoS issues have been focused and mainly addressed in the literature of MSNs. To the best of our knowledge, this paper is the first article to address QoE issues in emerging MSNs. In this paper, we first present a comprehensive investigation on recent advances in MSNs as well as QoE issues addressed in various types of applications and networks. From the lessons learned from the literature, then we propose a future research direction of QoE in MSNs.

• Access to Document (DOI)

On RoboCup@Home - Past, Present and Future of a Scientific Competition for Service Robots.

Dirk Holz, Javier Ruiz-del-Solar, Komei Sugiura, Sven Wachsmuth

RoboCup    686 - 697 2014

• Access to Document (DOI)

On RoboCup@ Home–past, present and future of a scientific competition for service robots

D Holz, J Ruiz-del-Solar, K Sugiura, S Wachsmuth

Robot Soccer World Cup, 686-697  2014

Non-monologue HMM-based speech synthesis for service robots: A cloud robotics approach

K Sugiura, Y Shiga, H Kawai, T Misu, C Hori

2014 IEEE International Conference on Robotics and Automation (ICRA), 2237-2242    2237 - 2242 2014

• Access to Document (DOI)

rospeex : A Cloud-based Spoken Language Communication Toolkit for ROS

SUGIURA Komei, HORI Chiori, ZETTSU Koji

電子情報通信学会技術研究報告 = IEICE technical report : 信学技報 (The Institute of Electronics, Information and Communication Engineers)  113 ( 248 ) 7 - 10 2013.10

ISSN  0913-5685

　View Summary

Stand-alone speech recognition/synthesis engines deployed on robots have drawbacks in accuracy and quality compared with cloud-based speech engines. In this paper, we propose "rospeex"', a novel cloud-based speech communication toolkit for ROS (Robot Operating System). It provides robots with high-performance speech recognition/synthesis capabilities with easy-to-use APIs. We have released the cloud-based service, so that researchers can use it without any cost/authentication.

階層型評価構造に基づく観光スポット推薦システムの構築と長期実証実験

杉浦孔明， 岩橋直人， 芳賀麻誉美， 堀智織

観光情報学 第 8 回研究発表会講演論文集, 9-12  2013

Development of robocup@ home simulation towards long-term large scale hri

T Inamura, JTC Tan, K Sugiura, T Nagai, H Okada

Robot Soccer World Cup, 672-680    672 - 680 2013

• Access to Document (DOI)

Complementary Integration of Heterogeneous Crowd-Sourced Datasets for Enhanced Social Analytics

R Lee, KS Kim, K Sugiura, K Zettsu, Y Kidawara

2013 IEEE 14th International Conference on Mobile Data Management 2, 234-243    234 - 243 2013

• Access to Document (DOI)

Human-robot interaction between virtual and real worlds: Motivation from robocup@ home

JTC Tan, T Inamura, K Sugiura, T Nagai, H Okada

International Conference on Social Robotics, 239-248    239 - 248 2013

• Access to Document (DOI)

ロボカップ@ ホームにおける日用品マニピュレーション

杉浦孔明， 長井隆行

日本ロボット学会誌 31 (4), 370-375 (The Robotics Society of Japan)  31 ( 4 ) 370 - 375 2013

ISSN  02891824

• Access to Document (DOI)

サービスロボットのための非モノローグ HMM による音声合成

杉浦孔明， 志賀芳則， 河井恒， 翠輝久， 堀智織

第 31 回ロボット学会学術講演会資料, pp. 2C1–02  2013

Utterance Classification Using Linguistic and Non-linguistic Information for Network-Based Speech-to-Speech Translation Systems

K Sugiura, R Lee, H Kashioka, K Zettsu, Y Kidawara

2013 IEEE 14th International Conference on Mobile Data Management 2, 212-216    212 - 216 2013

• Access to Document (DOI)

rospeex: クラウド型音声コミュニケーションを実現する ROS 向けツールキット

杉浦孔明， 堀智織， 是津耕司

信学技報 (CNR2013-10) 113, 7-10  2013

Development of Dialogue Systems "'Kyo-no Hanna' and 'Kyo no Osusume'"

翠 輝久, 水上 悦雄, 杉浦 孔明

情報通信研究機構季報 (情報通信研究機構)  58 ( 3 ) 25 - 29 2012.09

ISSN  1349-3191

DP-1-1 RoboCup@Home : A Benchmark Test for Domestic Service Robots

Sugiura Komei

Proceedings of the IEICE General Conference (The Institute of Electronics, Information and Communication Engineers)  2012 ( 2 ) "SS - 21"-"SS-22" 2012.03

観光案内への音声対話システムの活用

柏岡秀紀， 翠輝久， 水上悦雄， 杉浦孔明， 岩橋直人， 堀智織

デジタルプラクティス 3 (4), 254-261 3 ( 4 ) 254 - 261 2012

　View Summary

本論では，情報通信研究機構で開発，公開しているスマートフォン用音声対話システムAssisTra，観光スポット推薦システム京のおすすめについて述べる．AssisTraの公開に先立ちシステムの実稼働性，有効性，ユーザビリティ等の検証，問題点の洗い出し，ユーザの行動データ収集を目的として延べ100名を対象に実験を行った．その概要と結果についても報告する．さらに，京のおすすめについては，システム構築のために4000人のアンケートを実施した．その結果を基に，ユーザが問いかけやすい観点を見つけている．代表的な観点として「気分」「体験」「雰囲気」「特徴」に着目することにより，観光スポットを推薦するシステムである．

3-4 対話処理システム (はんなのガイド・京のおすすめ) の開発

翠輝久， 水上悦雄， 杉浦孔明， 岩橋直人

情報通信研究機構研究報告 58 (3.4), 25-29 (National Institute of Information and Communications Technology)  58 ( 3 ) 25 - 29 2012

ISSN  2187-767X

• Access to Document (DOI)

拡張モバイルマニピュレーションのための新規物体の学習

中村友昭， 杉浦孔明， 長井隆行， 岩橋直人， 戸田智基， 岡田浩之， ...

日本ロボット学会誌 30 (2), 213-224 (The Robotics Society of Japan)  30 ( 2 ) 213 - 224 2012

ISSN  02891824

　View Summary

It is convenient for users to teach novel objects to a domestic service robot with a simple procedure. In this paper, we propose a method for learning the images and names of these objects shown by the users. The object images are segmented out from cluttered scenes by using motion attention. Phoneme recognition and voice conversion are used for the speech recognition and synthesis of the object names that are out of vocabulary. In the experiments conducted with 120 everyday objects, we have obtained an accuracy of 91% for object recognition and an accuracy of 82% for word recognition. Furthermore, we have implemented the proposed method on a physical robot, DiGORO, and evaluated its performance by using RoboCup@Home's &ldquo;Supermarket&rdquo; task. The results have shown that DiGORO has outperformed the highest score obtained in the RoboCup@Home 2009 competition.

• Access to Document (DOI)