Deepgram ports Nova-3 speech model to AWS Trainium2 for batch inference
Overview
Deepgram says it ported its Nova-3 speech-to-text model to AWS Trainium2 for offline batch workloads, running eight model replicas on one Amazon EC2 trn2.3xlarge instance.
In an AWS Machine Learning Blog post, Deepgram reports that this setup processed 11 to 70 percent more audio per dollar on its throughput-oriented batch workloads than its previous compute options. The comparison uses instance pricing at the time of Deepgram's evaluation.
Written by AI from the articles below · updated Oct 19, 12:31 PM ET
Check the sources:
Article timeline
The articles in this story. Times are ET.
- AWS Machine Learning BlogOfficialDeepgram optimizes its Nova-3 speech-to-text model on AWS Trainium2
AIDeepgram ported its Nova-3 speech-to-text model to AWS Trainium2 for offline batch workloads, running eight model replicas on one Amazon EC2 trn2.3xlarge instance. On its throughput-oriented batch workloads, that setup processed 11–70 percent more audio per dollar than Deepgram's previous compute options, measured at instance pricing at the time of evaluation.
Heat Trend
Not enough continuous observations to show a trend yet.