Genesys Cloud - Main

We're removing the Transcription Confidence Threshold setting from Speech and Text Analytics and setting the threshold to 0 for all organizations.

What this means:
✅ Complete transcripts with every transcribed word preserved
✅ No more ellipses (…) replacing low-confidence transcript segments
✅ Better visibility into every customer interaction
✅ The transcription confidence threshold is no longer configurable

If you've previously configured a custom transcription confidence threshold, there's nothing you need to do. Genesys Cloud will remove the setting automatically, and all transcripts will include every transcribed word.

As transcription accuracy has continued to improve, preserving the full transcript now delivers more value than filtering low-confidence segments. It also helps improve downstream AI capabilities like sentiment analysis, topic detection, interaction summaries, and more.

Hi @Liore Finkelstein,

I have a question on this change and possible downstream impact.

We are exploring the AI assisted evaluations in genesys for voice calls and we have noticed that the accuracy of the AI assisted evaluation has decreased a lot for us starting the week of July 20.

What is happening for us is that AI is selectin the N/A answer and we have the feedback: Genesys AI could not accurately answer this question.

As per the documentation on genesys website, the explanation of the AI scoring as is below.

• For AI Scoring questions, the default answer is selected when the interaction is flagged as low accuracy or when the model confidence score is below 70%. For more information, see Automated scoring options in evaluation forms.

Any possibility that this change might impact the AI scoring?

Thanks
Enrik