Telling a text-to-speech reader how to pronounce things?
I tried reading my self-written ebook using Windows text-to-speech and was quite impressed with how well it worked, even on made-up proper nouns. But it stumbled here and there. It correctly read 1,000km as "one thousand kilometres", for example, but 12m was read as one-two-m rather than "12 metres". Is there a way to embed the correct pronunciation for words like this, that the Calibre reader can use?
I know I could rewrite that example as 12 metres, but there are other cases where that's not an option. Trickiest one I've noticed was a place where it used the wrong word stress for 'record', pronouncing it as the noun rather than the verb. Is there a way to tell the text-to-speech reader how to pronounce tricky stuff correctly? I'll put it in if it's easy to do.
I know many TTS readers can be manually coded by the user with rules on how to pronounce unfamiliar words - can I embed that information in the epub so that the users don't have to?
|