Introducing Nova-3 Pharma! #1694
naomi-deepgram
announced in
News + Announcements
Replies: 1 comment
|
Thanks for shipping Pharma. Its drug-name accuracy is great for our medical dictation app, but we're consistently seeing about 1 second more latency than nova-3-medical on prerecorded REST with identical audio and params (median 1,256 ms vs 485 ms, slower in 33/35 paired rounds). Benchmark details and request IDs are here: https://github.com/orgs/deepgram/discussions/1709. Is this expected, or is there a faster configuration? We'd love to move back to Pharma once latency matches. |
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
A transcript can be mostly right and still get the one word that matters most wrong: the medication name.
That’s the problem Nova-3 Pharma was built to solve.
Introducing Nova-3 Pharma: the first speech-to-text model purpose-built for the pharmaceutical industry, optimized for drug names, pharmaceutical terminology, and medical vocabulary.
Nova-3 Pharma ranks #1 in drug name recognition and overall transcription accuracy across batch and streaming among the models evaluated.
🔹 Purpose-built for drug names, pharmaceutical terminology, and medical vocabulary
🔹 High-quality general transcription performance you expect from Nova-3
🔹Built for pharmacy IVRs, prescription refill automation, healthcare voice agents, clinical documentation
🔹 Streaming and batch transcription with hosted API and self-hosted deployment
Learn more: https://deepgram.com/learn/introducing-nova-3-pharma
output.mp4
All reactions