Issue metadata
Sign in to add a comment
|
Words surrounded by spans are read individually in stead of as part of a sentence.
Reported by
henk.a...@gmail.com,
May 31 2017
|
||||||||||||||||||||||
Issue descriptionMode: force_next Version: 59.0.3071.47 Reproduction Steps: 1.Open the following url, this is a local newsletter generated daily. http://service.nu.nl/optiext/optiextension.dll?ID=he3hrgjlqDuzG%2Br2fGwrbCK3FFwJHFeSETdyhpf%2B%2BvFvO7H6o8HAPECmznBy6cGRG0OR6RSschk7UeNthdykjDT9Ybddh3 2.Try reading using either search+r, or search+right arrow. By the time you get to the actual news items you will notice that readout gets slower. On inspecting the html it looks as if all emphasized words are surrounded by a span tag. ChromeVox seems to decide to only read one span at a time. Windows screenreaders don do this and read the entire line ignoring the spans. 3.TalkBack has the same issue. It makes this type of newsletter almost impossible to read on Android or Chrome OS.
,
Sep 7
,
Sep 28
,
Jan 2
Select-to-Speak solved this by generating "paragraphs" of words from spans in the same block element (see select_to_speak/paragraph_utils.js). ChromeVox and Talkback could do the same thing if desired. If adding that logic to each client is redundant, we could also consider adding some block/paragraph parent information to each node in the Automation API. |
|||||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||||
Comment 1 by pbath...@chromium.org
, Jul 6 2017Status: Untriaged (was: Unconfirmed)