Pulse Pro
Model: Pulse Pro · smallest.ai
—
Response time · transcription only
Smallest AI's Pulse beats NVIDIA Nemotron on every dataset in both the Open ASR (ESB) and WildASR benchmarks, averaging 4.99% vs 5.96% WER on ESB and 7.78% vs 11.52% on robustness. Nemotron ships as free, self-hosted open weights; Pulse trades a per-minute fee for a fully managed API with SOC 2, HIPAA, and GDPR compliance.
NVIDIA
Model: Nemotron
—
Response time · transcription only
Pulse Speech-to-Text
Real-time transcription with industry-leading accuracy and ~64ms time-to-first-token, built for live voice agents.
30+ Languages
Accurate transcription across 30+ languages, with native English performance ranked #2 on the Open ASR Leaderboard.
Enterprise-Grade Compliance
SOC 2 Type II, GDPR, ISO 27001, and HIPAA-ready — with a Business Associate Agreement available for healthcare deployments.
Features
Benchmarks
Benchmark
▼
WER % on the Hugging Face OpenASR (ESB) benchmark. Lower is better. Pulse leads on every dataset and on aggregate (4.99% vs 5.96%). Source: internal evaluation on HF OpenASR datasets.
Frequently
asked questions
Is NVIDIA Nemotron a hosted API like Pulse?
How does Pulse accuracy compare to NVIDIA Nemotron?
What does running Nemotron actually cost?
Can Pulse STT run on-premise?