AssemblyAI STT 3.7 Realtime Agent vs Zoom Scribe Pro v2.1

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
  • Scribe Pro v2.11.55% · 178ms

AssemblyAI STT 3.7 Realtime Agent is ahead on time to first text and final-text delay, and Zoom Scribe Pro v2.1 on both datasets together. The Pipecat Dataset and The Ocular Dataset are too close to call.

Last updated Methodology by Dileep Chagam

Who is better at what

STT 3.7 Realtime AgentScribe Pro v2.1
  • Overall WER2.41%2.35%
  • Pipecat Dataset WER1.68%1.55%
  • Ocular Dataset WER3.81%3.87%
  • Time to first text493ms793ms
  • Final-text delay47ms178ms

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 AgentScribe Pro v2.1
  • ELO rating10241001–1039 · rank 1–1610461029–1053 · rank 1–10
  • Expected win rate47%53%
  • Ocular turns WER · 173 draws of 19816 won9 won
  • Pipecat WER · 932 draws of 99930 won37 won
  • Semantic WER · 194 draws of 1982 won2 won
  • Phone line · 174 draws of 19812 won12 won
  • Call-center noise · 135 draws of 19715 won47 won
  • Babble noise · 126 draws of 19722 won49 won
  • TV noise · 141 draws of 19816 won41 won
  • Reverb · 159 draws of 19824 won15 won
  • Short replies · 62 draws of 705 won3 won
  • Entities · 48 draws of 501 won1 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 Scribe Pro v2.1 FAQ

  • Zoom Scribe Pro v2.1 is more accurate on the public clips. On the 1,000 public Pipecat clips, AssemblyAI STT 3.7 Realtime Agent scores 2.41% and Zoom Scribe Pro v2.1 2.35%, so Zoom Scribe Pro v2.1 makes fewer errors by 0.07 points. On the 300 licensed Ocular turns, AssemblyAI STT 3.7 Realtime Agent scores 1.68% and Zoom Scribe Pro v2.1 1.55%, with likely ranges of 1.47 to 1.90% and 1.35 to 1.75%. The ranges overlap, so the difference is too close to call. Word error rate counts incorrect, missing, and extra words, so lower is better.

  • Yes. AssemblyAI STT 3.7 Realtime Agent finishes the transcript 300ms sooner. AssemblyAI STT 3.7 Realtime Agent has the final transcript 493ms after the speaker stops and Zoom Scribe Pro v2.1 793ms, at the median. Both are measured on the same 206 streamed turns.

  • AssemblyAI STT 3.7 Realtime Agent finishes the transcript 300ms sooner, and the two are too close to call on real conversations. 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 Zoom Scribe Pro v2.1 after 3.87%, at the median.

Test your voice agent the same way.

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