At a high level, improving one industry doesn't necessarily mean degrading another. When we add domain-specific data, it works to expand the model's understanding of that domain's language when applicable, rather than replacing or biasing it away from other domains like financial services. Theoretically of course that a bias can be achieved, but we monitor our training data-sets and benchmark transcription quality across industries to ensure there are no regressions - so improvements in one area remain safe for others.
Our models are trained on broad, diverse datasets, and domain data is incorporated in a way that improves recognition when relevant, without negatively impacting other use cases.
Original Message:
Sent: 04-23-2026 19:54
From: Peter Stoltenberg
Subject: Now Generally Available: Our New Native Transcription Engine for English and Spanish dialects
Hi Liore,
One Question I have is the following: Can you explain the concept of how transcription enhancements improve specific industries, without impacting the success rate of transcription for other industries? We are financial service shop, and releases for health care are cool, but doesn't apply to me, so I can see the question being asked: how do I know if that a release won't negatively affect me, in a different industry??
I have some business partners who are trying to understand this concept from a Genesys perspective, in comparison to other vendors, particularly who have more "fine tuned models" for our conversations and for our industry (or so they say).
------------------------------
Peter Stoltenberg
Engineer
Original Message:
Sent: 04-23-2026 16:37
From: Liore Finkelstein
Subject: Now Generally Available: Our New Native Transcription Engine for English and Spanish dialects
Hi Suzanne,
You should see improvements for healthcare, although it's not called out as a dedicated healthcare release. This release introduces a new generation of our native transcription model, which improves accuracy broadly across domains - including healthcare use cases.
In addition, we previously released a model that incorporated more healthcare-specific training data. We also have another model update planned for June, which includes additional healthcare data as part of broader training improvements.
So while there isn't a standalone "healthcare model" release, these ongoing updates are continuously improving transcription quality in healthcare scenarios.
Happy to share more specifics if that would be useful.
------------------------------
Liore Finkelstein
Product Manager
Original Message:
Sent: 04-22-2026 16:08
From: Suzanne Mueller
Subject: Now Generally Available: Our New Native Transcription Engine for English and Spanish dialects
Liore, does this release include improved native transcription for Healthcare? Months ago they had demo'd this to the Healthcare Provider Community, but I've struggled to find an update on it since.
------------------------------
Suzanne Mueller
Contact Center Engineer Sr.
IHC Health Services, Inc
suzanne.mueller@imail.org
Original Message:
Sent: 03-09-2026 15:58
From: Liore Finkelstein
Subject: Now Generally Available: Our New Native Transcription Engine for English and Spanish dialects
We're excited to announce that our new and improved Genesys native Speech to Text transcription engine for English and Spanish dialects is now officially Generally Available!
This release represents a major step forward in transcription quality across voice interactions - and the best part:
There's no extra cost. No configuration required.
Customers using Genesys native transcription are automatically benefiting from the upgrade.
With this new engine, you can expect:
✅ Improved accuracy and reduced Word Error Rate
✅ Greater consistency powering downstream analytics like Sentiment Analysis, Topic Spotting, Empathy Analysis, AI Insights and Automated Scoring.
High-quality transcription is the foundation of Speech and Text Analytics and AI based insights. By strengthening that foundation, we're enabling more precise insights and automations.
We're excited for you to experience the impact - and would love to hear your feedback!
#AICopilot(Agent,Supervisor,Admin)
#WEM-Quality,WFM,Gamification,etc
------------------------------
Liore Finkelstein
Product Manager
------------------------------