12:50:44 RRSAgent has joined #rqtf 12:50:44 logging to https://www.w3.org/2021/08/25-rqtf-irc 12:50:46 RRSAgent, make logs public 12:50:49 Meeting: Accessible Platform Architectures Working Group Teleconference 12:50:49 Date: 25 August 2021 12:51:32 agenda+ Joint working group meetings and break-out sessions planned for TPAC 2021. 12:51:32 agenda+ Synchronization Accessibility User Requirements. 12:51:32 agenda+ Natural Language Interface Accessibility User Requirements. 12:51:32 agenda+ Accessibility of Remote Meetings. 12:51:32 agenda+ Miscellaneous topics. 12:51:40 zakim, clear agenda 12:51:40 agenda cleared 12:51:59 agenda+ Joint working group meetings and break-out sessions planned for TPAC 2021. 12:51:59 agenda+ Synchronization Accessibility User Requirements. 12:51:59 agenda+ Natural Language Interface Accessibility User Requirements. 12:51:59 agenda+ Accessibility of Remote Meetings. 12:51:59 agenda+ Miscellaneous topics. 12:52:06 chair: jasonjgw 12:52:10 present+ 12:54:47 JPaton has joined #rqtf 12:57:14 janina has joined #rqtf 12:59:18 present+ 13:00:20 SteveNoble has joined #rqtf 13:00:29 present+ 13:01:42 scott_h has joined #rqtf 13:03:17 XAUR has been published - thanks to Josh for all hos work! 13:03:19 zakim, next item 13:03:19 agendum 1 -- Joint working group meetings and break-out sessions planned for TPAC 2021. -- taken up [from jasonjgw] 13:03:25 present+ 13:04:52 Janina: TPAC scheduled is slightly changed over ;ast week 13:05:07 scott_h_ has joined #rqtf 13:05:26 present+ 13:06:19 janina: May want to consider a breakout session where we introduce all our work, or perhaps have targeted cross-meetings 13:06:32 becky has joined #rqtf 13:07:08 Raja has joined #rqtf 13:07:50 janina: the issue was that some may not want to attend an hour-long eeting when they are only interested in one publication 13:08:32 Joshue108 has joined #rqtf 13:10:04 jasonjgw: Perhaps automotive accessibility is one area for a breakout 13:10:43 Ted: Yes, would be supportive of a breakout on automotive accessibility 13:10:54 Present+ 13:11:51 Ted: sent some proposed language to APA list for a breakout 13:12:31 Joshue108: Will update the wiki to reflect this additional area 13:14:16 jasonjgw: Natural language doesn't have a home for a specific working group 13:14:45 If you scheduled it, which week? 13:15:26 John_Paton present+ 13:15:41 janina: Proposed breakout on remote meetings 13:16:47 judy: Having a breakout for remote meetings would be good 13:18:03 judy: We will push for a time that will make it able for Scott to attend 13:18:30 TPAC agenda update with Auto A11y breakout 13:19:42 jasonjgw: Looks like we have agreed to automotive, remote meetings, and synchronization meetup with timed text 13:19:51 jasonjgw: Any others? 13:20:56 zakim, next item 13:20:56 agendum 2 -- Synchronization Accessibility User Requirements. -- taken up [from jasonjgw] 13:21:08 q+ 13:21:27 jasonjgw: We have two new issues which has been opened 13:22:13 jasonjgw: One was a suggestion for a reference, and the other was an issue around ASR accuracy 13:22:59 janina: Would prefer to handle these issues before publication 13:23:32 q+ 13:23:35 ack jan 13:25:19 janina: Section on audio description - we may need to figure out the preferred descriptor as in video description 13:27:37 q+ to ask if this is something to ask of localisation since the terminology changes between countries 13:29:24 q- 13:29:34 janina: Could include some connection to text descriptions as well, which is one of the reasons why video description is perhaps better 13:29:43 q+ 13:30:13 Judy has joined #rqtf 13:30:14 ack Josh 13:31:26 Joshue108: The comments to the BBC paper was around timing and accuracy 13:33:46 Joshue108: Comment on ASR also around accuracy 13:33:57 q? 13:33:59 ack raj 13:35:14 Raja: Support elimination of discussion around accuracy 13:35:20 q+ 13:36:15 q+ 13:37:03 Raja: The use of subtitles is a common term, but there is an SDH acronym used 13:38:59 Judy: we are doing captions differently so that we have both ASR and human cart captioning as well 13:40:02 q? 13:40:57 [SAUR] Claims about ASR accuracy #229 13:41:01 https://github.com/w3c/apa/issues/229 13:42:11 q? 13:42:14 ack judy 13:42:23 ack jan 13:43:22 https://raw.githack.com/w3c/apa/main/saur/#captions-in-live-media 13:44:02 Steve: what is stated in the document is strongly grounded in the research - free of editorial opinion/commentary. Additional research would be welcome on this point. There may be additional references. 13:44:47 Caption/SDH latency sensitivity for those who listen and read versus those who read only has several studies backing it, including the BBC paper 13:44:53 Steve: the research indicates taht ASR is much more timely than human captioning. 13:44:55 I can find a couple more if that helps 13:45:40 what is new is the discussion ASR latency. But since ASR is not usable as primary access, I am uncertain if ASR discussion is worth it 13:46:08 Steve notes that synchronization of human-created captions can easily be achieved with prerecorded mateiral, but live enviornments raise issues of latency vs. accuracy. If we only consider latency, then ASR is clearly the better solution, which is why accuracy needs to be considered - revealing the necessity of the trade-off. 13:46:32 Steve notes continued improvements in ASR accuracy, as dcumented in research findings. 13:47:24 q+ 13:47:36 Steve further notes the large data sets available to contemporary machine learning systems (e.g., from smart speaker/speech-based agent usage in large populations). 13:48:25 Steve: ASR is limited in its ability to recognize speech patterns in specific individuals and specific accents. 13:48:49 https://developer.mozilla.org/en-US/docs/Web/Guide/Audio_and_video_delivery/Adding_captions_and_subtitles_to_HTML5_video 13:48:56 Steve considers it important to cite this research and to include the discussion in the document. 13:49:00 s/taht/that 13:50:18 q+ 13:50:39 https://verbit.ai/the-differences-between-subtitles-closed-captions-and-sdh/ 13:50:45 janina: We need to reference the best language in current use of "subtitles" vs "captions" 13:50:45 Joshua I am 13:51:25 +1 to Raja's suggestion that SDH is likely widely recognised 13:51:30 q+ 13:52:56 ack Raja 13:53:31 Raja: ASR is improving, but still problematic, especially over the phone 13:54:00 Raja: Don't believe that ASR is equivalent to a live human 13:54:19 ack Josh 13:54:22 https://raw.githack.com/w3c/apa/main/saur/#caption-synchronization 13:55:25 JPaton: In the UK, subtitling in more standard as the term 13:55:29 q+ 13:56:34 JPaton: Speech-to-text has different levels, such as having respeakers which provide better recognition results 13:56:38 q+ to request that we consider adding reference to quality metrics 13:56:40 q- 13:57:39 Respeaking adds a human-to-the-loop 13:57:58 When I use interpreters + asr, that's similar. 13:58:00 q+ 13:58:08 I can't use ASR by itself since it is so variable 13:58:34 Judy: The need for quality metrics for ASR is a question 13:58:44 ack judy 13:58:44 Judy, you wanted to request that we consider adding reference to quality metrics 13:59:03 JPaton: 3 levels of speech to text: ASR, respeaking with ASR and STTR (palantypists etc) 14:00:23 John to contribute some language around the 3 levels 14:00:53 Metrics works for a single environment. For multiple environments, wow do you measure the variability? 14:01:21 zakim, bye 14:01:21 leaving. As of this point the attendees have been jasonjgw, scott_h, Joshue, janina, SteveNoble, Judy, scott_h_ 14:01:21 Zakim has left #rqtf 14:01:26 rrsagent, make minutes 14:01:26 I have made the request to generate https://www.w3.org/2021/08/25-rqtf-minutes.html jasonjgw 14:01:42 janina has left #rqtf 15:47:24 becky has joined #rqtf