question

timikoponen-8044 avatar image
0 Votes"
timikoponen-8044 asked SebastianMller-8520 edited

Ignoring the default lexicon

Hello, I'm making an app in which I would like to ignore the default lexicon in the text-to-speech service.

Testing with the ssml tag <lexicon uri="string"/>, I was only able to add more items to the lexicon and not completely replace it with an empty one.
https://docs.microsoft.com/en-us/azure/cognitive-services/speech-service/speech-synthesis-markup?tabs=csharp

Also the <say-as interpret-as="characters"> does not accomplish what I'm after.

In a more detail, when I'm using the finnish voice "fi-FI-NooraNeural" and using the text-to-speech to read out "ma", it is spoken out as "maanantai" (monday). But I would like it to be read just as "ma" without the lexicon replacing the default abbreviations.

Is there a way to disable the lexicon or to completely override it? Either disabling the lexicon as a property or with some option in the ssml tag, both options would work.

Thanks!

azure-speech
5 |1600 characters needed characters left characters exceeded

Up to 10 attachments (including images) can be used with a maximum of 3.0 MiB each and 30.0 MiB total.

1 Answer

GiftA-MSFT avatar image
0 Votes"
GiftA-MSFT answered

Hi, I don't think there's a way to remove the default lexicon. However, when you add custom lexicon, it would replace the pronunciation. Also, can you try adding an alias such as "maa", does it change to the desired pronunciation?

5 |1600 characters needed characters left characters exceeded

Up to 10 attachments (including images) can be used with a maximum of 3.0 MiB each and 30.0 MiB total.