From b789fa113d5b24d357e8bccdb47e5aa6acb52229 Mon Sep 17 00:00:00 2001 From: Fuqiao Xue Date: Mon, 24 Nov 2025 16:11:21 +0800 Subject: [PATCH] Fix typos and improve grammar in ruby-tts.md Corrected typos and grammatical errors in the document. --- explainers/ruby-tts.md | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/explainers/ruby-tts.md b/explainers/ruby-tts.md index 83a9a2b..2d1e1f9 100644 --- a/explainers/ruby-tts.md +++ b/explainers/ruby-tts.md @@ -194,7 +194,7 @@ but not always. 一郎, a given name, can be Ichirō or Kazuo… * Names written in ideographs can occasionally be assigned non-standard readings through ruby, - replacing the standard reading(s) that the reader would other otherwise have assumed. + replacing the standard reading(s) that the reader would otherwise have assumed. Therefore, attempting to determine pronunciation only from the base text is not fully reliable either. @@ -258,7 +258,7 @@ Its values are: whether to behave as `complementary` or as `phonetic`. In all cases, it is understood that the above requirements -describe the default behavior speech user agents. +describe the default behavior of speech user agents. It is always acceptable to enable users to opt into alternative behaviors via a variety of settings, (e.g. A person Bilingual in Chinese and Japanese but more familiar with Chinese culture @@ -343,7 +343,7 @@ Generalizing from [the earlier description](#the-ruby-type-attribute): If the annotations and the base disagree, the annotations take precedence. If the annotations disagree with each other, - the choice is somewhat arbitray, + the choice is somewhat arbitrary, but we propose that the outermost annotation takes precedence. * if multiple annotations on the same base have `ruby-type=auto`, we propose to let the user agent decide for itself which is which. @@ -366,7 +366,7 @@ Generalizing from [the earlier description](#the-ruby-type-attribute): similarly to how it would be had the annotation been inserted as an inline parenthetical. * Morphological or contextual analysis is often used to determine the correct - pronunciation of text based on its surronding, + pronunciation of text based on its surrounding, including un-annotated text. As substituting the annotation for the base or concatenating both could confuse the analysis, @@ -405,7 +405,7 @@ Generalizing from [the earlier description](#the-ruby-type-attribute): > Issue: How do we solve this? > Options include: > 1. `ruby-type=phonetic` adds a reading for the annotated word - > in the ua's pronunciation dictionary to be used on later unannotated words, + > in the UA's pronunciation dictionary to be used on later unannotated words, > but the UA is still expected to perform morphological and contextual analysis > to pick the right reading. > Seem appropriate and flexible, @@ -567,7 +567,7 @@ The downsides of this proposal are: * `base` as a value is not terribly useful, and arguably a foot-gun. If an annotation is present, it generally conveys something, and hiding that from the user is likely unwise. - More over, in exceptional cases, + Moreover, in exceptional cases, if there really is a need, this can already be achieved by adding `aria-hidden` on the annotation. * Although it could seems that `annotation` would be the right value to reach for