Deepgram Boosts Nova‑3 Speech‑to‑Text Performance on AWS Trainium2, Cutting Costs
Current overview (not historical)
Deepgram has ported and optimized its Nova‑3 speech‑to‑text model to run on AWS Trainium2, achieving a 11%–70% increase in audio processed per dollar on a single device. The team used TorchNeuron to migrate the PyTorch model, then applied the NKI Lib and custom kernels to raise throughput. They also built a Rust‑based data‑parallel runtime to eliminate Python overhead, enabling the entire model to run on Trainium2 without falling back to CPU. These optimizations demonstrate that specialized hardware can significantly reduce operating costs for large‑scale voice AI workloads, while maintaining or improving performance compared to traditional CPU‑based deployments.
Written with AI from the reporting · updated 10/19/2026
EventGraph Explorer
One time cutoff for reports, relations and evidence.
New at this step ·Deepgram Boosts Speech‑to‑Text Efficiency 11‑70% Using AWS Trainium2
- Reports in replay
- 1
- First-party
- 1
- Sources
- 1
Reports first seen at this step · 1
- AWS Machine Learning Blog · First-partyDeepgram Boosts Speech‑to‑Text Efficiency 11‑70% Using AWS Trainium2Original source ↗
Discovery-time replay; the linked graph and evidence share this cutoff. The full overview and activity remain current.
Event developments
1 developments10/20/2026, 1:11:41 AM
Deepgram Boosts Speech‑to‑Text Efficiency 11‑70% Using AWS Trainium2AWS Machine Learning Blog
Filtered by first discovery; only representative reports already discovered by this cutoff are shown. Story grouping may have changed later, so this is not a complete historical snapshot.
Relations at this time
Loading time-scoped relations…
Bounded reconstruction, not an exact snapshot: entity names, relationship text and story grouping may include later revisions. Overview and full source archive below remain current.
Story route
A station-by-station route from the first public evidence onward; every station carries its own evidence and uncertainty.
- 1Oct 20, 00:00recorded Oct 20, 01:11single source
Deepgram 将 Nova-3 语音模型优化至 AWS Trainium2 实现成本降低
Read original source ↗ AWS Machine Learning BlogReplay this moment in the explorer
Uncertainty marks flag where to check the sources; they are not proven causation.
Full source archive (current)
Current public reports; outside the historical replay.
First-party sources · 1 reports
- AWS Machine Learning Blog · Oct 20, 00:23Deepgram Boosts Speech‑to‑Text Efficiency 11‑70% Using AWS Trainium2
Deepgram ported its Nova‑3 speech‑to‑text model to AWS Trainium2, deploying eight replicas on a single trn2.3xlarge EC2 instance. The optimized configuration processed 11–70 % more audio per dollar on batch transcription workloads compared to previous compute options. The team used TorchNeuron for native PyTorch integration, performed kernel‑level tuning, and built a Rust‑based data‑parallel runtime to eliminate Python overhead. These gains enable Deepgram to serve more audio for the same infrastructure budget and lower costs for customers who self‑host on SageMaker AI.
Read original source ↗
Independent coverage · 0 reports
No verifiable reports in this group yet.
Different reports do not necessarily disagree. Check original sources and publication dates.
Report timeline
Explore every side through dated source reports.
- AWS Machine Learning BlogSelectedDeepgram Boosts Speech‑to‑Text Efficiency 11‑70% Using AWS Trainium2
Deepgram ported its Nova‑3 speech‑to‑text model to AWS Trainium2, deploying eight replicas on a single trn2.3xlarge EC2 instance. The optimized configuration processed 11–70 % more audio per dollar on batch transcription workloads compared to previous compute options. The team used TorchNeuron for native PyTorch integration, performed kernel‑level tuning, and built a Rust‑based data‑parallel runtime to eliminate Python overhead. These gains enable Deepgram to serve more audio for the same infrastructure budget and lower costs for customers who self‑host on SageMaker AI.