AssemblyAI STT 3.7 Realtime Agent vs Gradium

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)
  • STT 3.7 Realtime Agent1.68% · 47ms
  • Gradium4.02% · 219ms

AssemblyAI STT 3.7 Realtime Agent is ahead of Gradium 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

STT 3.7 Realtime AgentGradium
  • Overall WER2.41%3.83%
  • Pipecat Dataset WER1.68%4.02%
  • Ocular Dataset WER3.81%3.47%
  • Time to first text493ms1.78s
  • Final-text delay47ms219ms

Lower is better on every measure. Too close to call means the likely ranges overlap.

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.

STT 3.7 Realtime AgentGradium
  • ELO rating10241001–1039 · rank 1–16978973–983 · rank 15–21
  • Expected win rate57%43%
  • Ocular turns WER · 171 draws of 19816 won11 won
  • Pipecat WER · 765 draws of 999195 won39 won
  • Semantic WER · 190 draws of 1983 won5 won
  • Phone line · 159 draws of 19825 won14 won
  • Call-center noise · 108 draws of 19778 won11 won
  • Babble noise · 100 draws of 19791 won6 won
  • TV noise · 147 draws of 19740 won10 won
  • Reverb · 152 draws of 19836 won10 won
  • Short replies · 63 draws of 704 won3 won
  • Entities · 47 draws of 503 won0 won

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

STT 3.7 Realtime Agent vs Gradium FAQ

  • AssemblyAI STT 3.7 Realtime Agent is more accurate on both datasets. On the 1,000 public Pipecat clips, AssemblyAI STT 3.7 Realtime Agent scores 2.41% and Gradium 3.83%, so AssemblyAI STT 3.7 Realtime Agent makes fewer errors by 1.42 points. On the 300 licensed Ocular turns, AssemblyAI STT 3.7 Realtime Agent scores 1.68% and Gradium 4.02%, so AssemblyAI STT 3.7 Realtime Agent makes fewer errors by 2.35 points. Word error rate counts incorrect, missing, and extra words, so lower is better.

  • Yes. AssemblyAI STT 3.7 Realtime Agent finishes the transcript 1.28s sooner. AssemblyAI STT 3.7 Realtime Agent has the final transcript 493ms after the speaker stops and Gradium 1.78s, at the median. Both are measured on the same 206 streamed turns.

  • AssemblyAI STT 3.7 Realtime Agent is the stronger pick for voice agents: it makes fewer errors on real conversations and finishes the transcript 1.28s 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 STT 3.7 Realtime Agent shows first text after 3.81% and Gradium after 3.47%, at the median.

Test your voice agent the same way.

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