Idea: Content-Sensitive Transcription Formatting
We all know that WHEN SOMEONE TYPES IN ALL-CAPS it’s interpreted as shouting. So why shouldn’t it work the other way, too? Now that everyone has devices capable of transcription in their pockets, maybe we can work on making transcription easier by using vocal tones for formatting cues.
IF I SHOUT A SENTENCE INTO MY PHONE, IT SHOULD BE TRANSCRIBED IN ALL CAPS.
If I only shout one word of a sentence, it should be italicized, especially if it’s an adverb.
Someone must be working on more seamless formatting of transcription, right? Nuance, the company which makes the speech recognition engine behind Siri, has a feature in their own software that they call Natural Punctuation, but even that’s limited to just automatic periods and commas.
Could we have a setting that automatically interprets upspeak as an indication to end a sentence with a question mark?
What other kinds of formatting could be managed easily by changing your inflection rather than your content?