Correcting Pronunciation of TTS Audio

Post a reply

Smilies
:D :) ;) :( :o :shock: :? 8-) :lol: :x :P :oops: :cry: :evil: :twisted: :roll: :!: :?: :idea: :arrow: :| :mrgreen: :geek: :ugeek:
BBCode is ON
[img] is ON
[flash] is OFF
[url] is ON
Smilies are ON
Topic review
   

Expand view Topic review: Correcting Pronunciation of TTS Audio

Re: Correcting Pronunciation of TTS Audio

Post by Ernie » Wed Feb 12, 2025 6:22 am

I would recommend trying to fake-spell a work, like "jobe". Siri seems to say it like you want, with long-o. Amazon won't say it like you want. If going this route, this means continuing to put TTS text in [ ] and trying to fake the TTS sound with alternate spellings.

"about possibly adding in IPA or X-SAMPA tags for pronunciation did that get implemented?"
Yes. so...

If using Amazon Polly, you can define the sound using IPA (international phonic alphabet). But it may be tricky to figure out how to use the write characters to get the sound you want. In this case, you don't use [ ], but need to put HTML that Amazon will then use to create the TTS sound. Example text:

You say, <phoneme alphabet="ipa" ph="pɪˈkɑːn">Pecan</phoneme>

In this case, Amazon will ignore the "Pecan", and instead sound it like "pɪˈkɑːn".

These are are the only options I have for you that I can think of.

Re: Correcting Pronunciation of TTS Audio

Post by abishur » Mon Feb 10, 2025 10:48 am

I found this post from 2019

viewtopic.php?f=2&t=5448&p=14835&hilit=phonetic&sid=75871341dcddf399c2bb0aae4eaf93b3#p14835

about possibly adding in IPA or X-SAMPA tags for pronunciation did that get implemented?

Correcting Pronunciation of TTS Audio

Post by abishur » Thu Feb 06, 2025 10:46 am

Is there a way to instruct the TTS audio how to pronounce words? I saw that I can use [ ] to change what is read verse what is shown. So for example if I had a card that said 444-425 BC and it read that as "Four hundred forty four dash four hundred twenty five BC" I could enter the card as:
Code: Select all
444-425 BC [Four forty four to four twenty five BC]

And get it to read it the way I want it to.
But what if I have the name Job (that's with a long O) and it read's it as job (i.e. going to work with a short o). How do I explicitly enter the phonetic pronunciation of the word?

Top