09:00:09 <RRSAgent> RRSAgent has joined #smartagents-main
09:00:13 <RRSAgent> logging to https://www.w3.org/2026/02/26-smartagents-main-irc
09:00:16 <zohar> zohar has joined #smartagents-main
09:00:53 <plh> Meeting: Smart Voice Agents - Day 2
09:02:43 <chiaradm> chiaradm has joined #smartagents-main
09:03:00 <plh> Present+ IK Ngoo, GinaSmith, DeborahDahl, ZoharGan, PaoloaDiMaio, Emmet Coin, Ulrike Stiefelhagen, ChiaraDeMartin, GerardChollet, KazuyukiAshimura, plh
09:04:54 <kaz> kaz has joined #smartagents-main
09:05:21 <Zakim> Zakim has joined #smartagents-main
09:05:23 <plh> present+ BenjaminWeiss, DirkSchnelle-Walka
09:05:45 <plh> present+ ShadiAbou-Zahra
09:05:45 <kaz> meeting: W3C Workshop on
09:05:46 <kaz> Smart Voice Agents - Session 2
09:05:53 <dirk> dirk has joined #smartagents-main
09:05:55 <kaz> zakim, who is on the call?
09:05:55 <Zakim> Present: BenjaminWeiss, DirkSchnelle-Walka, ShadiAbou-Zahra
09:06:27 <plh> present+ IK Ngoo, GinaSmith, DeborahDahl, ZoharGan, PaoloaDiMaio, Emmet Coin, Ulrike Stiefelhagen, ChiaraDeMartin, GerardChollet, KazuyukiAshimura, plh
09:06:34 <plh> zakim, who's here?
09:06:34 <Zakim> Present: BenjaminWeiss, DirkSchnelle-Walka, ShadiAbou-Zahra, IK, Ngoo, GinaSmith, DeborahDahl, ZoharGan, PaoloaDiMaio, Emmet, Coin, Ulrike, Stiefelhagen, ChiaraDeMartin,
09:06:38 <Zakim> ... GerardChollet, KazuyukiAshimura, plh
09:06:38 <Zakim> On IRC I see dirk, Zakim, kaz, chiaradm, zohar, RRSAgent, bkardell, ddahl, plh
09:07:56 <kaz> present+ EmmetCoin
09:08:12 <kaz> present+ EmmettCoin
09:08:17 <kaz> present- EmmetCoin
09:08:24 <kaz> present- Emmet
09:08:30 <kaz> present- Coin
09:09:26 <kaz> topic: Scene setting
09:09:50 <kaz> dd: (gives explanations on the goals, logistics and expectations)
09:10:43 <kaz> topic: Accessibility of 3D and Immersive Content via Voice Interaction = Zohar Gan
09:10:50 <kaz> s/=/-/
09:10:53 <kaz> rrsagent, make log public
09:10:58 <kaz> rrsagent, draft minutes
09:10:59 <RRSAgent> I have made the request to generate https://www.w3.org/2026/02/26-smartagents-main-minutes.html kaz
09:13:58 <kaz> zg: (describes difficulties and challenges around accessibility for 3D/Immersive Content)
09:14:54 <plh> present+ KristiinaJokinen
09:15:56 <chiaradm> chiaradm has joined #smartagents-main
09:17:17 <chiaradm> chiaradm has joined #smartagents-main
09:18:03 <chiaradm> present+
09:18:08 <chiaradm> chiaradm has left #smartagents-main
09:28:19 <kaz> kaz: great input for existing W3C groups around media handling and web-based digital twins
09:28:27 <kaz> ... suggest you work with those groups
09:28:40 <kaz> ... can send details about those groups to you later
09:28:50 <kaz> zg: great
09:29:03 <kaz> ec: this is great and beneficial
09:29:07 <kaz> zg: tx!
09:30:50 <kaz> pdm: quick question
09:31:00 <kaz> ... tx for your presentation
09:32:14 <kaz> ... consideration to natural voice
09:32:40 <kaz> ... whether the developers are thinking of making more natural voice for the system
09:33:08 <kaz> present+ YashGhelani
09:33:27 <kaz> ... naturalness of the voice
09:33:36 <kaz> zg: so would make the voice more natural?
09:33:39 <kaz> pdm: yes
09:33:50 <kaz> zg: doing poc
09:33:58 <kaz> ... good idea to explore
09:34:07 <plh> present+ YashGhelani
09:34:32 <kaz> ... previous solution was using recording of people's voice itself
09:34:51 <kaz> ... another approach is using Web API for more natural voice and customization
09:34:55 <kaz> ... maybe we can use that
09:35:08 <kaz> ... from the practical viewpoint, can be fast enough
09:35:15 <kaz> pdm: tx
09:35:27 <kaz> ... advances of natural speech
09:35:33 <kaz> ... so just want to suggest
09:35:49 <kaz> ... UX would be better with more natural speech
09:36:26 <kaz> topic: Transition of Use Cases for Voice to LLM-based RAG or Agent setups in difficult scenarios - Ulrike Stiefelhagen
09:37:24 <kaz> us: (present about hallucinations in Voice agents)
09:38:10 <kaz> ... (describes several use cases)
09:38:49 <kaz> ... (existing difficulties with voice interface)
09:40:08 <kaz> ... (mentions patients' Voice Assistant "Juki")
09:41:36 <kaz> ... (pros and cons of voice interface)
09:46:35 <kaz> ... (opportunity of GenAI)
09:47:08 <kaz> ... (another use case of Workers' Voice Assistant "Helping Harry"
09:47:23 <kaz> ... (within noisy environment)
09:49:23 <kaz> s/GenAI/GenAI and Voice/
09:49:29 <kaz> rrsgent, draft minutes
09:51:02 <kaz> ... (opportunity for GenAI & Voice again)
09:51:34 <kaz> ... (deal with different pronunciations)
09:52:03 <kaz> s/rrsgent, draft minutes//
09:52:09 <kaz> rrsagent, draft minutes
09:52:10 <RRSAgent> I have made the request to generate https://www.w3.org/2026/02/26-smartagents-main-minutes.html kaz
09:53:09 <kaz> ec: years ago working on a similar system for warehouse
09:53:20 <kaz> ... the question was demanding timing
09:53:34 <kaz> ... GenAI model can handle that?
09:53:40 <kaz> ur: didn't use picking
09:53:52 <kaz> ... we chose workstations build up things
09:54:12 <kaz> ... regarding the speed/timing, we also see difficulty
09:54:25 <kaz> ... don't generate timing
09:54:35 <kaz> ... it's basically done ahead
09:54:44 <kaz> ... don't need variation at the moment
09:55:37 <kaz> ... estimate the timing for our work
09:55:50 <kaz> ... traffic light says time needed
09:56:17 <kaz> ... each of work step depends of fraction of time
09:56:36 <kaz> ... 150 steps around clock
09:56:50 <kaz> ... there is some flexibility
09:57:55 <kaz> kaz: worked on realtime OS based speech timing for my Ph.D thesis 10 years ago
09:58:28 <kaz> ... it depends on use cases but strict timing management would be useful for more natural dialogue-based communication with GenAI
09:58:40 <kaz> ... would suggest we work on that kind of advanced use cases as  welll
09:59:21 <kaz> topic: Towards Smarter Voice Interfaces: Using Grounding and Knowledge - Kristiina Jokinen
09:59:32 <kaz> dd: can play your video for you
09:59:52 <kaz> kj: sorry about my bad connection today
10:00:10 <kaz> dd: (starts Kristiina's recorded video)
10:00:14 <plh> present+ Jazmin
10:00:23 <kaz> rrsagent, draft minutes
10:00:24 <RRSAgent> I have made the request to generate https://www.w3.org/2026/02/26-smartagents-main-minutes.html kaz
10:01:06 <kaz> kj: (gives her talk about smarter voice interfaces)
10:01:51 <kaz> ... (challenges for voice interaction)
10:05:11 <kaz> ... (Errors in voice interaction)
10:06:33 <kaz> ... (Grounding: Collaborative mechanism)
10:10:24 <kaz> ... (Knowledge graphs for grounding)
10:11:56 <kaz> ... (Agentic Architecture)
10:15:40 <kaz> dd: tx!
10:15:50 <kaz> ... do you think you can get questions?
10:15:53 <kaz> kj: let's try!
10:16:06 <kaz> ec: 2 questions
10:16:14 <kaz> ... formalism for grounding structure?
10:16:24 <kaz> ... knowledge graphs and DB?
10:16:36 <kaz> ... passed from one AI agent to another?
10:16:49 <kaz> kj: very important point
10:16:58 <kaz> ... so far we have basically worked for different projects
10:17:13 <kaz> ... the expertise from those projects could be applied
10:17:23 <kaz> ... also explore what we can do
10:17:39 <kaz> ... it's very important to try to work on some formalization
10:17:59 <kaz> ... so that other can easily build their systems
10:18:28 <kaz> ... should dive into a bit more details
10:18:37 <kaz> ... about how to build the systems
10:18:55 <kaz> ... so far we've had kind of dialog modelling aspect
10:19:15 <kaz> ... but then make it computable needs more discussion
10:19:33 <kaz> ec: another simple question
10:20:07 <kaz> ... do you address the interaction, e.g., about the date using different styles?
10:20:12 <kaz> kj: good question
10:20:27 <kaz> ... same entity with different expressions within the knowledge graph
10:21:07 <kaz> ... if you actually can refer to the entity which is already grounded, the description could be shorten
10:21:16 <kaz> ec: tx, interesting
10:21:22 <kaz> rrsagent, draft minutes
10:21:23 <RRSAgent> I have made the request to generate https://www.w3.org/2026/02/26-smartagents-main-minutes.html kaz
10:21:54 <kaz> topic: Towards Web Standards for Configurable Naturally Responsive Voice Interaction for AI Agents - Paola Di Maio
10:22:03 <kaz> dd: Paola has issue with connection
10:22:18 <kaz> ... maybe Gerard could show her recorded video
10:22:23 <kaz> gc: let me try
10:22:43 <kaz> ... (shows Paola's recorded Video)
10:22:54 <kaz> ... wondering about the sound
10:22:58 <kaz> dd: try again
10:26:13 <kaz> pdm: (traditional pipeline for speech-to-speech: STT -> LLM -> TTS)
10:27:20 <kaz> ... (great tech but missing UX)
10:30:12 <kaz> ... (7 critical usability failures)
10:32:50 <kaz> ... (proposed UX requirements)
10:38:50 <kaz> ... (then shows a demo)
10:39:34 <kaz> rrsagent, draft minutes
10:39:36 <RRSAgent> I have made the request to generate https://www.w3.org/2026/02/26-smartagents-main-minutes.html kaz
10:43:15 <kaz> ec: any work around speech recognition
10:43:37 <kaz> ... to detect whether the speech is finished or not
10:43:59 <kaz> pdm: very technical
10:44:23 <kaz> ... there is some work done
10:46:10 <kaz> kaz: when I was a ASR researcher 10 years ago, worked on a kind of big speech corpus to detect the readiness of utterance for interactive dialog
10:46:40 <kaz> ... maybe we might think about even bigger dialog corpus for today's use cases from research viewpoint
10:47:02 <kaz> pdm: yeah, many things to be done for research
10:47:28 <kaz> kj: did you just use the information for end-to-end interaction?
10:47:49 <kaz> pdm: there is a draft paper as the basis of the demo
10:47:56 <kaz> ... 5 use cases there
10:48:12 <kaz> ... each use case use specific script
10:48:25 <kaz> ... note that models are changing every moment
10:48:40 <kaz> ... variables related to the users also change
10:48:53 <kaz> ... I have an outline
10:50:00 <kaz> dd: some more discussion would be great during the breakout sessions
10:51:25 <kaz> topic: Gaze-Aware Dialog Systems - Fares Abawi
10:51:50 <kaz> gc: (shows recorded video for Fares)
10:52:04 <kaz> ... (Under utilized modality)
10:52:18 <kaz> ... (where gaze matters)
10:54:23 <kaz> ... (multimodal pipeline)
10:54:35 <kaz> ... (multimodal fusion)
10:55:57 <kaz> ... (neural integration)
10:56:37 <PaolaDM> PaolaDM has joined #smartagents-main
10:57:02 <PaolaDM> Thanks for setting up the IRC channel
10:59:49 <kaz> ... (standardization needs)
11:03:13 <kaz> pdm: wondering what kind/level of role "gaze" would play
11:03:36 <kaz> dd: a use case is trend users
11:03:58 <kaz> pdm: how much train the users for the gaze modality is a question
11:04:38 <kaz> kaz: reminded us of the W3C MMI WG's work
11:04:49 <kaz> ... maybe we can talk wit him about that too
11:04:53 <kaz> dd: yeah
11:05:25 <kaz> us: in the factory context, we can add a context using gaze modality
11:05:43 <kaz> kj: very interesting work
11:05:53 <kaz> ... a lot of research around gaze and turn taking
11:06:03 <kaz> ... it's kind of point device
11:06:19 <kaz> ... wondering if there is some way to distinguish things from each other
11:06:30 <kaz> ... when we want to take turns
11:06:39 <kaz> ... implying our expectations
11:06:50 <kaz> rrsagent, draft minutes
11:06:52 <RRSAgent> I have made the request to generate https://www.w3.org/2026/02/26-smartagents-main-minutes.html kaz
11:08:22 <kaz> dirk: useful for automotive environment
11:08:53 <kaz> dd: time for wrapping up the presentation part
11:09:06 <kaz> ... and then moving to the breakout part
11:09:16 <kaz> [10-min break]
11:09:21 <kaz> rrsagent, draft minutes
11:09:22 <RRSAgent> I have made the request to generate https://www.w3.org/2026/02/26-smartagents-main-minutes.html kaz
11:22:07 <kaz> topic: Breakouts
11:23:21 <kaz> zakim, who is here?
11:23:21 <Zakim> Present: BenjaminWeiss, DirkSchnelle-Walka, ShadiAbou-Zahra, IK, Ngoo, GinaSmith, DeborahDahl, ZoharGan, PaoloaDiMaio, Ulrike, Stiefelhagen, ChiaraDeMartin, GerardChollet,
11:23:24 <Zakim> ... KazuyukiAshimura, plh, EmmettCoin, KristiinaJokinen, chiaradm, YashGhelani, Jazmin
11:23:24 <Zakim> On IRC I see PaolaDM, dirk, Zakim, kaz, RRSAgent, bkardell, ddahl, plh
11:23:40 <kaz> dirk: (gives instructions for breakouts)
11:24:32 <kaz> ... suggestion from Debbie was combining some of the talks
11:25:24 <kaz> [[
11:25:26 <kaz> Room 1 - Kristiina
11:25:26 <kaz> Room 2 - Zohar
11:25:26 <kaz> Room 4 - Ulrike
11:25:26 <kaz> Room 5 - Paola/Fares
11:25:27 <kaz> ]]
11:26:12 <kaz> dirk: slides on guidance and notes
11:26:15 <kaz> [[
11:26:39 <kaz> Group A (Kristiina Jokinen, Zoom Room 1): https://docs.google.com/presentation/d/1nKTrSX0VmyC1dNyF5e9JegX5Ggwwst4UnlQVR8eL3E8/edit?usp=sharing
11:26:40 <kaz> Group B (Zohar Gan, Zoom Room 2): https://docs.google.com/presentation/d/1wYd6U3OCj2fFhdKobsRJkWzRPY5knLajxi8o0Csg-8M/edit?usp=sharing
11:26:40 <kaz> Group C (Fares Abawi, Paola Di Maio, Zoom Room 3): https://docs.google.com/presentation/d/1paQE60SVoe5xmFAHA1cvVmJuVXmdGuMYd0iOGeHwWAI/edit?usp=sharing
11:26:40 <kaz> Group D (Ulrike Stiefelhagen, Zoom Room 4): https://docs.google.com/presentation/d/1X7Nb5s5fWzd21SX5x5m8-erFFu9FAWNTzGXdDFUCCU4/edit?usp=sharing
11:26:42 <kaz> ]]
11:27:31 <kaz> dd: dirk has sent an email about how to join the breakout zoom
11:27:36 <kaz> dirk: did that
11:28:35 <kaz> plh: btw, this main room will become the Room 1
11:28:50 <kaz> rrsagent, draft minutes
11:28:51 <RRSAgent> I have made the request to generate https://www.w3.org/2026/02/26-smartagents-main-minutes.html kaz
11:30:35 <zohar> zohar has joined #smartagents-main
11:31:58 <zohar> I see a message in room 2 saying the host has another meeting in progress and I can't join
11:32:32 <ddahl> @plh zoom says the host has another meeting in progress when I try to join room 2
11:32:42 <plh> hu...
11:32:50 <plh> I'm in room 2
11:37:27 <plh> only room 1 and 2 are now running. not enough participants in room 3 and 4.
11:57:25 <kaz> topic: People coming back to the main room
11:58:39 <kaz> s/topic: People coming back to the main room/[People coming back to the main room]/
11:58:44 <kaz> rrsagent, draft minutes
11:58:45 <RRSAgent> I have made the request to generate https://www.w3.org/2026/02/26-smartagents-main-minutes.html kaz
12:05:00 <kaz> present+ FaresAbawi
12:05:24 <kaz> (discussion about gase, and also turn taking)
12:07:40 <kaz> bw: not only gaze but various signals could be used
12:09:02 <kaz> ... that's kind of like playing a boardgame
12:09:21 <kaz> fa: detecting tension directed to tasks is needed
12:09:45 <kaz> ... the model should detect that
12:10:00 <kaz> ... e.g., my gaze indicating something or sometimes not
12:10:21 <kaz> ... we can have different multiple stages
12:10:38 <kaz> ... and the model can ignore some of them depending on the situation
12:10:57 <kaz> ... you can't expect the model can see everything
12:11:06 <kaz> ... the model has no idea about our intention
12:11:39 <kaz> ... there could be extensions based on our intention like smart glasses
12:11:46 <kaz> ... there are many open questions
12:11:58 <kaz> ... need formalization
12:12:17 <kaz> ... I'm more concerned about operation of users
12:12:54 <kaz> ... if it's very functional, there still could be problems with requiring specific Web application
12:13:07 <kaz> kj: we need to define scenarios
12:13:25 <kaz> ... and need to know if the gaze means something around user's interaction
12:13:59 <kaz> ... if the gaze helps the system to identify the intention, can be kind of monitor
12:14:10 <kaz> ... but there is another aspect
12:14:23 <kaz> ... what your gaze tells for people
12:14:36 <kaz> ... providing some intention to people and the system
12:14:50 <kaz> ... so that they can manage the situation better
12:15:57 <kaz> ... reminded of some work on standardization
12:16:13 <kaz> ... when we built in some technology or system
12:16:34 <kaz> ... some kind of standardization can be achieved in a sense of technology
12:16:58 <kaz> ... useful and helpful but how to measure the usefulness?
12:18:06 <kaz> dd: time to wrap up. the other group was talking about accessibility of 3D and Immersive content
12:18:29 <kaz> zg: media semantic meta data
12:18:52 <kaz> ... important part of the data
12:19:14 <kaz> ... helpful to improve privacy and latency using hybrid approach
12:19:52 <kaz> ... producing semantic metadata for accessibility
12:20:03 <kaz> ... collaboration within W3C would be useful
12:20:22 <kaz> ... WebVTT, WebVMT, etc.
12:20:51 <kaz> ... someone's gaze to identify the intention
12:20:55 <kaz> dd: tx
12:21:16 <kaz> ... we can continue the plenary discussion here at the main channel
12:21:38 <kaz> topic: Closing
12:21:48 <kaz> dd: one comment on gaze
12:22:07 <kaz> ... gaze in the real world is difficult to handle
12:22:17 <kaz> ... much bigger than the screen
12:22:34 <kaz> ec: what do you look at is a question
12:22:53 <kaz> fa: as long as you know about the screen
12:23:11 <kaz> ... but completely different with the real world situation
12:23:24 <kaz> ... but quite good way to segment objects
12:23:31 <kaz> ... useful when you talk about entities
12:23:51 <kaz> ... gazing upon something is already a question to be detected
12:24:04 <kaz> ... in a 3D world, segmenting object is useful
12:24:49 <kaz> ... maybe someone else can talk about more
12:24:53 <kaz> rrsagent, draft minutes
12:24:54 <RRSAgent> I have made the request to generate https://www.w3.org/2026/02/26-smartagents-main-minutes.html kaz
12:25:03 <kaz> kj: segmentation is related to time information
12:25:43 <kaz> ... probably want to know what kind of gaze pattern is being used
12:25:52 <kaz> ... how difficult to detect it
12:26:08 <kaz> ... time-based segmentation is important
12:26:14 <kaz> fa: definitely
12:26:33 <kaz> ... just wanted to mention the importance within the real world
12:28:26 <kaz> kaz: probably we need to think about possible use cases for synchronization of ...
12:28:36 <kaz> ec: yes, there are synchronization problems
12:29:03 <kaz> ... how do we synchronize multiple date?
12:29:10 <kaz> ... not just humans vs AI
12:29:18 <kaz> ... need to think about different time scale
12:29:36 <kaz> kj: need to think about some storage
12:29:50 <kaz> ... helpful memory
12:30:08 <kaz> ... maybe sometimes we once go back to some point
12:30:34 <kaz> ec: maybe we need to enforce AI to wait
12:30:57 <kaz> ... would be helpful more natural conversations
12:31:46 <kaz> kj: if there are many people, the conversation tend to split into several sub groups
12:32:16 <kaz> dd: we have to make sure whatever depending on the culture
12:32:21 <kaz> ... one culture and another
12:32:42 <kaz> ... get used to different cultures
12:32:53 <kaz> ... dangerous to assume one specific culture
12:33:00 <kaz> ec: yeah
12:33:20 <kaz> dd: people has different immigration history
12:33:25 <kaz> ... even in US
12:34:18 <kaz> s/of .../of multiple data streams/
12:34:41 <kaz> kj: coaching for elderly adults
12:34:56 <kaz> ... poc experiment to evaluate usefulness
12:35:06 <kaz> ... what topics people wanted to talk about
12:35:18 <kaz> ... and see how the system should react
12:35:40 <kaz> ... it's community-oriented
12:36:23 <kaz> ... in the JP context, the system can know the usage
12:36:40 <kaz> ... in the sort of western context, systems are kind of tools for information
12:36:57 <kaz> ... emotional understanding of intelligence
12:37:13 <kaz> ... it was interesting when you want to decide the system
12:37:18 <kaz> ... can be one for all
12:37:39 <kaz> ... but personalization and adoption would be useful
12:38:17 <kaz> ... working in different cultures would require different viewpoints
12:38:50 <kaz> pdm: not only culture but there is individual difference as well
12:38:56 <kaz> ... a lot of circumstances thre
12:38:59 <kaz> s/thre/there/
12:39:13 <kaz> dd: cultural difference based on cultures
12:39:37 <kaz> fa: collective cultures
12:40:03 <kaz> ... usually would look at a small cluster depending on some specific culture
12:40:18 <kaz> ... or wide glance for the whole people
12:40:38 <kaz> ec: sometimes we use "gaze" to focus on something
12:40:50 <kaz> ... carries several meanings
12:41:11 <kaz> ... we can detect reliably the standard usage of "gaze"
12:41:26 <kaz> us: the difference between "gaze" and "gesture"
12:41:41 <kaz> ... how we look
12:41:52 <kaz> ... how to track gaze?
12:41:59 <kaz> fa: we use web cam
12:42:13 <kaz> ... something called web gaze for browsers
12:42:21 <kaz> ... first step is calibration
12:42:42 <kaz> ... you don't force the users to purchase special devices
12:42:54 <kaz> s/web gaze/web gazer/
12:43:09 <kaz> ec: web gazer?
12:43:18 <kaz> plh: https://webgazer.cs.brown.edu/
12:43:30 <kaz> rrsagent, draft minutes
12:43:31 <RRSAgent> I have made the request to generate https://www.w3.org/2026/02/26-smartagents-main-minutes.html kaz
12:43:53 <kaz> fa: this is more like a poc
12:44:14 <kaz> us: could use gaze tracking
12:44:27 <kaz> ... and the target of the gaze is a square
12:44:34 <kaz> fa: exactly
12:44:50 <kaz> pdm: interact with voice?
12:45:09 <kaz> ... if people has difficulty with speaking
12:45:20 <kaz> ... how to use gaze for voice?
12:45:45 <kaz> fa: first focus on 3 things
12:45:50 <kaz> ... turn-taking
12:45:57 <kaz> ... divergent
12:46:06 <kaz> ... and pointing
12:46:12 <kaz> rrsagent, draft mintues
12:46:12 <RRSAgent> I'm logging. I don't understand 'draft mintues', kaz.  Try /msg RRSAgent help
12:46:19 <kaz> s/rrsagent, draft mintues//
12:46:23 <kaz> rrsagent, draft minutes
12:46:24 <RRSAgent> I have made the request to generate https://www.w3.org/2026/02/26-smartagents-main-minutes.html kaz
12:46:36 <kaz> ... that's my proposal essentially
12:46:59 <kaz> topic: Wrap-up
12:47:08 <kaz> dd: a lot of interest in "gaze"
12:47:23 <kaz> ... for example, for turn-taking and divergence
12:47:41 <kaz> ... combination with voice for interaction
12:47:51 <kaz> ... sounds like a recommendation from this group
12:47:58 <kaz> ... anything else?
12:49:25 <kaz> kaz: integration of multiple modalities/resources including "gaze" is important
12:49:54 <kaz> ... also as Emmett mentioned time synchronization among those resources would be important for advanced use cases
12:49:57 <kaz> dd: right
12:50:05 <kaz> ... regarding tomorrow's session
12:50:30 <kaz> -> https://www.w3.org/2025/10/smartagents-workshop/agenda.html#session3 Session 3
12:50:43 <kaz> dd: 4 topics
12:50:52 <kaz> ... real-time processing
12:51:01 <kaz> ... in-vehicle interaction
12:51:08 <kaz> ... trust/empathy
12:51:20 <kaz> ... beyond screen readers
12:51:33 <kaz> dirk: tx a lot for your presentations!
12:51:53 <kaz> kj: thank you very much from me too
12:52:00 <kaz> rrsagent, draft minutes
12:52:01 <RRSAgent> I have made the request to generate https://www.w3.org/2026/02/26-smartagents-main-minutes.html kaz
12:52:50 <kaz> dd: if you won't be able to join tomorrow's session, all the presentations will be published later
12:52:57 <kaz> kj: appreciated
12:53:38 <kaz> [Session 2 adjourned]
12:53:42 <kaz> rrsagent, draft minutes
12:53:43 <RRSAgent> I have made the request to generate https://www.w3.org/2026/02/26-smartagents-main-minutes.html kaz
13:25:36 <kim> kim has joined #smartagents-main
14:12:30 <Zakim> Zakim has left #smartagents-main
15:41:49 <kim> kim has joined #smartagents-main
15:42:58 <kim> kim has left #smartagents-main