Template:TTS

<onlyinclude><{{{html|span}}} class="tts {{{class|}}}" {{#if:{{{lang|}}}|lang="{{{lang}}}"|}} {{#switch: {{{ParameterWhichShouldAlwaysBeEmpty|}}}|{{{tts|}}}{{{2|}}}=title{{=}}""|{{{tts|}}}=title{{=}}"{{{2|}}}"|{{{2|}}}=title{{=}}"{{{tts|}}}"}}><span class="visual" aria-hidden="true">{{{1}}}</span><span class="help"> (pronounced </span><i class="aural">{{#switch: {{{tts|}}} | {{{tts}}} | {{{2|error: TTS pronunciation missing}}} }}</i><span class="help">)</span></{{{html|span}}}></onlyinclude>

Demo

displayed string (pronounced error: TTS pronunciation missing)

displayed string in older style (pronounced displayed string in older style, plus this text only displayed to TTS)

oddly-pronounced word demo: 1337 (pronounced error: TTS pronunciation missing)

Description

This is a template for giving bits of text a TTS pronunciation.

Earlier, aria-label was mistakenly used for this, but it turns out that text substitution for speaking-machines is something that is not currently well-supported in HTML. (This can be a problem when it comes to the temptation to turn everything into "AI" — understanding that every piece of text written by humans is made of arbitrary signifiers, it is incredibly tempting to simply build a dynamic signifier-interpreting agent that actively learns all the exceptions to every rule instead of properly building a system where "dumb" programs or bits of data can easily pass along less-ambiguous messages to each other.)