benchmarks/stt by coval/openai vs deepgram
speech-to-text head-to-head · data by Coval

OpenAI vs Deepgram: Speech-to-Textlatency & accuracy

Measured head-to-head, not marketing: the verdict is split Deepgram is faster (98 ms median TTFS vs 632 ms), while OpenAI is more accurate (2.6% WER vs 4.5%). Independent data, refreshed daily, on identical inputs. Voice quality and price are not measured.

OpenAI or Deepgram — which is better?

Each provider's best model on each measured axis, last 7d. Lower is better on both:

AxisOpenAI (best model)Deepgram (best model)Measured winner
Speed — median TTFSgpt-4o-mini-transcribe · 632 msnova-3 · 98 msDeepgram
Accuracy — avg WERgpt-4o-mini-transcribe · 2.6%nova-3 · 4.5%OpenAI

Not measured here: price — treat those as vendor claims until measured. Full leaderboard context: the independent speech-to-text benchmark →

Every OpenAI and Deepgram model, measured

All benchmarked models from both providers on the same pinned dataset, ranked fastest-first (median TTFS), last 7d:

ModelProviderTTFS medianTTFS p95WERSamples
nova-3Deepgram98 ms154 ms4.5%3,327
nova-2Deepgram98 ms175 ms5.2%3,330
gpt-4o-mini-transcribeOpenAI632 ms1,236 ms2.6%3,347
gpt-realtime-whisperOpenAI666 ms821 ms3.3%3,347
gpt-4o-transcribeOpenAI767 ms1,158 ms3.1%3,353

Independent data, by Coval

Numbers are from the voice benchmark by Coval, a voice-AI evaluation platform that does not sell STT models — it measures every provider as a neutral third party on a pinned dataset under production-realistic conditions. Openbenchmarks mirrors the results daily with attribution; methodology and runner are open-source and reproducible.

synced from Coval2026-07-13 13:40 UTC · full STT benchmark → · methodology →

OpenAI vs Deepgram — common questions

Is OpenAI faster than Deepgram for speech-to-text?

On the current 7d window, OpenAI's fastest model (gpt-4o-mini-transcribe) has a median TTFS of 632 ms, vs 98 ms for Deepgram's fastest (nova-3) — so Deepgram is faster on measured median latency. Distributions matter too: p95 is 1,236 ms for OpenAI vs 154 ms for Deepgram.

Which is more accurate, OpenAI or Deepgram?

By Word Error Rate: OpenAI's best model (gpt-4o-mini-transcribe) averages 2.6%, vs 4.5% for Deepgram's best (nova-3) — OpenAI leads on measured accuracy. Lower is better; WER is measured on identical inputs under production-realistic conditions.

Which should I pick — OpenAI or Deepgram?

The measured verdict is split: Deepgram is faster (median TTFS), OpenAI is more accurate (WER). Pick by your constraint — real-time voice agents care about latency; fidelity-critical workloads care about WER. Note price is not measured here.

Where does this data come from?

The independent voice benchmark by Coval, a voice-AI evaluation platform that does not sell STT models. Every model runs the same pinned dataset under production-realistic conditions, re-run roughly every 30 minutes; Openbenchmarks mirrors the results daily with attribution. Methodology and runner are open-source (Apache-2.0).