t-room 1 2 2 2 2 1 1 2 t-room 2 Development of Assistant System for Ensemble in t-room Yosuke Irie, 1 Shigemi Aoyagi, 2 Toshihiro Takada, 2 Keiji Hirata, 2 Katsuhiko Kaji, 2 Shigeru Katagiri 1 and Miho Ohsaki 1 Because of the importance of synchronization, a synchronized collaborative work can be a good measure for evaluating the utility of a network-based remote collaboration assistant system, such as t-room, which inevitably suffers from delay in communication. Focusing on this, we study how one can perform synchronized works in t-room by conducting a remote session for playing musical instruments. In place of conventional devices such as canon, polyrhythm, and troll, we attempt to reduce the adverse effect of delay by using a metronome and the idea of local lag. Through experimental music performances, we clarify the effects of metronome and local lag for reducing the difficulty in remote sessions suffering from delay. 1. Fig. 1 1 t-room An example of t-room installation. t-room 1 1) t-room 1 Graduate School of Engineering, Doshisha University 2 NTT NTT Communication Science Laboratories, Nippon Telegraph and Telephone Corporetion 1 c 2009 Information Processing Society of Japan
2) 2 2 t-room 2 t-room 1 / t-room 2 250ms 120bpm 8 3)4)5)6)7)8) t-room 2 Fig. 2 Ensemble with telecommunication delay. 2. 2) 2.1 3 2.2 2 c 2009 Information Processing Society of Japan
2) 4 9) 1 2.3 1 2 Fig. 5 5 Synchronized transmission of metronome sound by using local lag. 5 3. 3.1 6 Fedora Core 8 OS 1 PC PC t-room 1 t-room 1 4 3 4 LAN WAN Fig. 3 Schematic diagram of local lag. Fig. 4 Schematic diagram of extended local lag. 3 c 2009 Information Processing Society of Japan
Fig. 6 6 Overview of proposed ensemble assistant system. Fig. 7 7 Schematic diagram of network delay measurement. 8 Fig. 8 Schematic diagram of data flow controlled by local lag. t-room t-room CPUIntel(R) Core(TM) 2 CPU 2.66 GHz 3GB OSFedora Core 8 M-AUDIO Delta 1010LT Linear-16 44.1 khz 2 MIDI CD 44.1 khz 3.2 TCP UDP TCP UDP 2 2 t-room 7 t 0 TCP. t 1. t 2 (t 2 t 1 ) t 3 t 3 t 0 t 2 t 1 2. 3.3 4 c 2009 Information Processing Society of Japan
2 2 8 8.. 10) 9 3.4 10 10 4. 4.1 t-room WAN Fig. 9 9 Schematic diagram of local lag control. Fig. 11 10 Fig. 10 An example sound of local lag feedback. 11 Experimental settings of remote ensemble. LAN Dummynet 11) 11 PC Dummynet PC t-room PC 5 c 2009 Information Processing Society of Japan
100BASE-TX t-room Dummynet LAN t-room 4.2 1 1 12) 8 8 1 5 4 4 8 2 Dummynet 10ms 0ms10ms 90ms100ms 11 11 1 1 2 3 3 3 33 1 5 1 4.3 12 13 12 Fig. 12 Average opinion scores for accompaniment part. 13 Fig. 13 Average opinion scores for melody part. 12 13 14 15 14 1 15 1 2 1 2 3 1 2 4 0 4.4 12 13 6 c 2009 Information Processing Society of Japan
15 14 Fig. 14 Average misalignment for trained players. Fig. 15 Average misalignment for untrained players. 4 3 2 4 3 1 1 1 2 12 15 3 40ms 50ms 2 2 1 14 15 1 2 14 15 0 15 5. 7 c 2009 Information Processing Society of Japan
2 4 1), t-room, NTT, Vol.19, No.6 (2007 6 ), pp.10-12(2007) 2) Dane Stuckel and Carl GutwinThe Effects of locallag on Tightly-Coupled Interaction in Distributed GroupwareComputer Supported Cooperative Workpp.447-456(2008). 3), Vol98No.74pp.95-102(1997). 4), OpenRemoteGIG Vol43No.2pp.299-309(2002). 5) GDS(global delayed session)music Vol2002No.63, pp.61-66, (2002). 6) Vol2002No.123pp.23-28(2002). 7),,, 2003 06 MVE,10/13(2003). 8) Vol2003No.127pp.43-48(2003). 9) Tom StaffordMatt WebMINDHACKSpp.175-176(2008) 10) pp.125-144(2004) 11) L. Rizzo. Dummynet: a simple approach to the evaluation of network protocols. ACM Computer Communication Review, Vol.27, No.1, pp.31-41(1997). 12) 8 c 2009 Information Processing Society of Japan