Recitation indicates that something is a link, footnote or similar (e.g. through a “pling”) which can be configured by the user.
Existing: No
To do: Recognise certain tags and play a sound.
Identified as a component during the pilot study.
Description
Description
Related Objects
Related Objects
Event Timeline
Comment Actions
The user should also be able to take action on "pling":
- Link: follow link
- Footnote: play footnote
Comment Actions
When T158954: Use XPath to get text nodes related to utterances is done, add some kind of objects to cleaned content for the tags that should trigger a notification. It may be preferable to focus on a link that should be followed, though this may cause problems if you rely on tabbing. Behaviour for footnote (and possibly other tags) will be different.
In order to have time to act on a notification, a grace period after the recitation of the triggering expression may be needed. This will probably cause issues for several links in a row.
Comment Actions
This will need to be split up into multiple parts:
- Notify on encountering tags
- Interact with encountered tags
- Interact with links (follow)
- Interact with footnotes (read footnote and return to current position)
Some needed decisions are:
- Where does the notification get added (in TTS server, before, after)
- What does the notification consist of: a non-verbal sound, a word or both (or configurable combination).
- "♪pling♪ link"
- "♪plong♪ footnote"
Comment Actions
This was mentioned in the feedback from SVWP test: https://sv.wikipedia.org/wiki/Wikipedia:Bybrunnen/Arkiv_2021-06#Testa_Wikispeech.