Need help with text analytics - building the dictionary for improving speech to text based on domain

[ganhaan]

Hey

We are trying to use speech & text analytics api for setting up dictionaries for proper transcription of words ( speech to text) based on finance domain. However we encountered issues with couple of words like CD or Zelle which most often get transcribed into something else leading to errors in our processes. Wanted to understand if there is any guidance on these items.

[Leor]

Hello! If you're referring to transcription, you can check out the post here:

New Dictionary Management API

Hi community! I'm Leor - one of the product manager's looking after voice transcription in Genesys Cloud. We recently launched Dictionary Management through API which allows for accuracy improvement in native voice transcription. You can find the documentation in API explorer here: The API allows for several inputs to the dictionary: dialect term examplePhrases phrases source boostValue soundsLike At a basic level, dictionary management (referred to in the API as dictionaryfeedback)…

Here are a few additional examples:

[ganhaan]

Thanks for sharing this examples. I have followed these examples for the terms CD and Zelle which common in financial institutions. How ever the genesis transcription engine still doesnt recognize them right. for zelle i have tried "soundsLike": [
"zel", "zell", "zele", "zelly"
]
and for CD have tried "soundsLike": [
"cd", "seedee", "c d"
]