Forum Discussion
AI Voice Generation emphasis in SL
Sorry for the quick delay here—I took this back to the team to see if anyone had thoughts/suggestions similar to what you shared about that break time markup. It seems like there's consensus that emphasis in particular is hard to achieve, when I think about this it makes sense because it's not quite a pronunciation difference, I can see why the speech models would have trouble with it! The feeling on the team is there's some experimentation needed to get the voice to flow correctly, and that sometimes experimentation with pronunciation can achieve close to what you want for emphasis.
I think you've probably already seen this based on what you referenced, but for anyone else following this thread who may be curious, here is an article the team put together that talks about some of the limitations and options with SSML models and AI speech.
Curious to keep following this and see if there are any specific practices folks have landed on that worked really well to achieve emphasis.
I don't know that I can agree this is solved; I did the quotes around the word and it helped but wasn't strong enough. I had tried the <emphasize...> option and that did not work at all (didn't see notes that it would, but thought I'd try). Hopefully will see the bold option in text/notes to make it easier to get the desired result without spending too much time fiddling with tags
As you say, quote marks help but they're not perfect. The only tags I've found that work at all are <break time="xs" /> to insert a pause. It can be frustrating at times to get the correct intonation.
I've recently taken to breaking text up into smaller chunks, so that it's quicker to regenerate if I'm not happy with the emphasis/pronunciation on a particular bit. It's more work than should be necessary, but I guess this is still an emerging technology.