14:46:17 <RRSAgent> RRSAgent has joined #epub-fxl
14:46:21 <RRSAgent> logging to https://www.w3.org/2024/02/05-epub-fxl-irc
14:46:21 <Zakim> RRSAgent, make logs Public
14:46:22 <Zakim> please title this meeting ("meeting: ..."), wendyreid
14:46:32 <wendyreid> meeting: FXL Accessibility Task Force
14:46:37 <wendyreid> date: 2024-02-05
14:46:42 <wendyreid> chair: wendyreid
14:59:31 <JonasLillqvist> JonasLillqvist has joined #epub-fxl
15:01:48 <Hadrien> Hadrien has joined #epub-fxl
15:02:16 <gpellegrino> gpellegrino has joined #epub-fxl
15:02:19 <Hadrien> present+
15:02:20 <jgriggs_prh> jgriggs_prh has joined #epub-fxl
15:02:25 <jgriggs_prh> present+
15:02:29 <gpellegrino> present+
15:04:25 <wendyreid> present+
15:04:59 <CharlesL> CharlesL has joined #epub-fxl
15:05:14 <gpellegrino> scribe+
15:05:33 <gpellegrino> wendyreid: I'm working on the final revision of the main document
15:05:39 <CharlesL> present+
15:06:28 <wendyreid> https://github.com/w3c/publishingcg/issues/69
15:06:45 <gpellegrino> ... I would like to speak about the issue Hadrien opened last week
15:07:28 <gautierchomel> gautierchomel has joined #epub-fxl
15:07:56 <gpellegrino> Koko: I work in Penguin Random House
15:08:49 <gpellegrino> wendyreid: while discussing about TTS we understood that there is no standard on top we can develop TTS engines
15:08:58 <Hadrien> q+
15:09:03 <wendyreid> ack Hadrien
15:09:33 <gpellegrino> Hadrien: the issue I created only some of the issues in creating a TTS engine
15:10:07 <gpellegrino> I left out highlighting, virtual highlighting, switching voices between languages, etc.
15:11:13 <gpellegrino> ... I think we need to document what we see in real files
15:12:01 <gpellegrino> wendyreid: when I looked at Speech API, I've seen that it is a CG Note, it's not a REC, it's not officiale
15:12:28 <Hadrien> q+
15:12:52 <wendyreid> ack Hadrien
15:12:56 <gpellegrino> ... I think there are issues about resources needed to create audio with good voices
15:13:07 <gpellegrino> Hadrien: in my experience it's wide supported
15:13:29 <gpellegrino> ... but it is quite inconsistent
15:14:49 <gpellegrino> ... its difficult to have feedback from the engines
15:15:17 <gpellegrino> ... actually you can use native APIs if you're developing native apps
15:16:21 <gpellegrino> ... I don't think the status of the document is an issue for us
15:16:41 <gpellegrino> ... because our work should be on what to do before passing data to Speech API
15:17:10 <gpellegrino> wendyreid: I prefer to work on stable specs
15:18:01 <gpellegrino> ... Hadrien is right, this work is about "pre-API" work
15:18:03 <gpellegrino> q+
15:18:11 <wendyreid> ack gpellegrino
15:18:13 <wendyreid> scribe+
15:18:39 <Hadrien> q+
15:18:39 <wendyreid> gpellegrino: So I understand the issue, the idea is to define a way to extract text from EPUBs in an abstract layer that then you can pass to the TTS
15:18:47 <wendyreid> ... or display to the end user as simple text
15:18:54 <wendyreid> ... meaning for FXL, we can have a reflow version
15:18:58 <wendyreid> ack Hadrien
15:19:10 <gpellegrino> Hadrien: It's more than text
15:19:23 <gpellegrino> ... it's also semantics: language + context
15:19:45 <gpellegrino> ... may also intercept hidden elements (like page breaks)
15:20:33 <gpellegrino> ... the output should be an object with the text plus metadata
15:21:02 <gpellegrino> wendyreid: a lot of TTS engines in reading apps do not read alt-text for images
15:21:40 <gpellegrino> Hadrien: in Readium alt-text is supported (no way to skip it, for the moment)
15:22:37 <gpellegrino> wendyreid: the base on which we can build this engine is to have a good semantic in the text
15:23:00 <gpellegrino> ... I'm making research on how browsers manage reading mode
15:23:35 <Hadrien> q+
15:23:40 <gpellegrino> ... I've found an article about bugs in the browser reading mode
15:23:57 <wendyreid> https://www.sarasoueidan.com/blog/tips-for-reader-modes/
15:24:02 <wendyreid> ack Hadrien
15:24:43 <gpellegrino> Hadrien: I'm less concerned about that, I think the biggest issue we may have with FXL is similar to liquid mode with a well known fixed format for digital publications
15:25:25 <gpellegrino> ... if the reading mode is correct, we may still have problems in text splitted in multiple spans
15:25:49 <gpellegrino> ... I think we should reconstruct the content
15:26:50 <gpellegrino> ... problems with semantics may raise up with content splitted in multiple pages
15:28:38 <gpellegrino> wendyreid: I think that we may have a pause across different pages (in the middle of a paragraph)
15:29:23 <gpellegrino> Hadrien: if we creare a object for the whole publication, and we reconstruct the content using ICU tokenizer, we may be able to recreate sentences
15:29:54 <gpellegrino> wendyreid: we may have performance issues
15:30:27 <gpellegrino> Hadrien: it depends, we may have different approaches
15:31:54 <gpellegrino> wendyreid: In FXL we may use the TOC to reconstruct part of the content
15:34:21 <gpellegrino> Hadrien: it's not that simple, since you have things splitted in multiple pages
15:34:36 <gpellegrino> wendyreid: maybe we can benefit from using DPUB ARIA roles
15:35:18 <gpellegrino> Hadrien: sure, maybe also having something to say where a sentence ends
15:37:24 <gpellegrino> ... I think documenting what developers are already doing
15:37:40 <gpellegrino> ... is a good starting point for creating this document
15:37:56 <gpellegrino> wendyreid: which question may we ask?
15:38:19 <gpellegrino> Hadrien: I think we can start with asking how do they break the content (based on HTML, etc.)
15:38:35 <gpellegrino> ... and knowing how do they manage non textual content
15:38:40 <gpellegrino> with semantics
15:38:42 <gpellegrino> q+
15:39:29 <gpellegrino> ... I think this system may also be useful for remediation process
15:39:42 <wendyreid> ack gpellegrino
15:39:52 <wendyreid> gpellegrino: It might be worth asking about MathML
15:40:02 <wendyreid> ... TTS breaks on MathML, might also be interesting
15:40:16 <wendyreid> ... reader mode and TTS, we might need multiple approaches to MathML
15:40:20 <CharlesL> q+
15:40:29 <wendyreid> ack CharlesL
15:40:53 <CharlesL> cant find the window
15:42:15 <gpellegrino> CharlesL: will we have FXL ebooks with MathML?
15:42:40 <CharlesL> q+
15:42:40 <gpellegrino> gpellegrino: I think the scope of this document is also for reflowable
15:42:48 <wendyreid> ack CharlesL
15:43:03 <Hadrien> q+
15:43:12 <gpellegrino> wendyreid: I think we can mention MathML without speaking about the format
15:43:28 <wendyreid> ack Hadrien
15:43:37 <gpellegrino> CharlesL: I think there should be an option for reading "invisible" things like alt-text
15:44:04 <gpellegrino> Hadrien: sure, maybe we should then discuss about skippability and escapability
15:45:11 <gpellegrino> wendyreid: I think another important element is personalization
15:47:08 <gpellegrino> Hadrien: we have a call on Readium community this Wednesday, I may add an agenda item about this
15:47:45 <gpellegrino> ... for example Readium mobile and Readium desktop have two different approaches and we may start documenting them
15:48:10 <CharlesL> s/cant find the window//
15:48:35 <gpellegrino> wendyreid: AOB?
15:49:59 <gpellegrino> Hadrien: the more I look at it, the more it seems to me useful for different usecases
15:51:10 <CharlesL> CharlesL has left #epub-fxl
15:52:50 <wendyreid> zakim, end meeting
15:52:50 <Zakim> As of this point the attendees have been Hadrien, jgriggs_prh, gpellegrino, wendyreid, CharlesL
15:52:53 <Zakim> RRSAgent, please draft minutes
15:52:54 <RRSAgent> I have made the request to generate https://www.w3.org/2024/02/05-epub-fxl-minutes.html Zakim
15:53:01 <Zakim> I am happy to have been of service, wendyreid; please remember to excuse RRSAgent.  Goodbye
15:53:01 <Zakim> Zakim has left #epub-fxl
17:00:55 <wendyreid> rrsagent, bye
17:00:55 <RRSAgent> I see no action items