AssemblyAI Universal 3.6 Pro vs Deepgram Nova-3

Compare

Accuracy and speed

Left gets fewer words wrong. Lower has the final text ready sooner.

Accurate and fast0ms200ms400ms600ms800ms0%2%4%6%Words wrong on clean audio (Pipecat WER)
  • Universal 3.6 Pro1.77% · 91ms
  • Nova-33.46% · 103ms

AssemblyAI Universal 3.6 Pro is ahead of Deepgram Nova-3 on both datasets together, the Pipecat Dataset, time to first text and final-text delay. The Ocular Dataset is too close to call.

Last updated Methodology by Dileep Chagam

Who is better at what

Universal 3.6 ProNova-3
  • Overall WER2.71%3.78%
  • Pipecat Dataset WER1.77%3.46%
  • Ocular Dataset WER4.50%4.39%
  • Time to first text485ms1.06s
  • Final-text delay91ms103ms

Lower is better on every measure. Too close to call means the likely ranges overlap. For costs, see our breakdown of AssemblyAI pricing.

Clip by clip

On every clip both models were scored on, the lower WER wins, and WERs within 5 points of each other are a draw; in the rating, a wider win counts for more. Expected win rate is what the two ELO ratings predict for a match.

Universal 3.6 ProNova-3
  • ELO rating10211003–1031 · rank 2–16971940–1015 · rank 5–22
  • Expected win rate57%43%
  • Ocular turns WER · 146 draws of 19631 won19 won
  • Pipecat WER · 805 draws of 999154 won40 won
  • Semantic WER · 189 draws of 1964 won3 won
  • Phone line · 154 draws of 19830 won14 won
  • Call-center noise · 111 draws of 19767 won19 won
  • Babble noise · 101 draws of 19886 won11 won
  • TV noise · 121 draws of 19856 won21 won
  • Reverb · 142 draws of 19843 won13 won
  • Short replies · 50 draws of 6814 won4 won
  • Entities · 43 draws of 504 won3 won

Short replies are decided on the error count and entities on exact capture, with no draw margin. How the rating is worked out

Universal 3.6 Pro vs Nova-3 FAQ

  • AssemblyAI Universal 3.6 Pro is more accurate on both datasets. On the 1,000 public Pipecat clips, AssemblyAI Universal 3.6 Pro scores 2.71% and Deepgram Nova-3 3.78%, so AssemblyAI Universal 3.6 Pro makes fewer errors by 1.07 points. On the 300 licensed Ocular turns, AssemblyAI Universal 3.6 Pro scores 1.77% and Deepgram Nova-3 3.46%, so AssemblyAI Universal 3.6 Pro makes fewer errors by 1.69 points. Word error rate counts incorrect, missing, and extra words, so lower is better.

  • Yes. AssemblyAI Universal 3.6 Pro finishes the transcript 579ms sooner. AssemblyAI Universal 3.6 Pro has the final transcript 485ms after the speaker stops and Deepgram Nova-3 1.06s, at the median. Both are measured on the same 206 streamed turns.

  • AssemblyAI Universal 3.6 Pro is the stronger pick for voice agents: it makes fewer errors on real conversations and finishes the transcript 579ms sooner. A voice agent cannot reply until the final transcript arrives, and the Ocular recordings are real conversations.

  • Yes, both return partial text before the speaker stops. AssemblyAI Universal 3.6 Pro shows first text after 4.50% and Deepgram Nova-3 after 4.39%, at the median.

Test your voice agent the same way.

Cekura runs these simulated calls on your agent and scores every one.