AssemblyAI STT 3.7 Realtime Agent vs Google Chirp 2

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
  • Chirp 22.59% · 568ms

AssemblyAI STT 3.7 Realtime Agent is ahead of Google Chirp 2 on both datasets together, the Pipecat Dataset, time to first text and final-text delay. The Ocular Dataset is too close to call. Google Chirp 2 has a newer release: AssemblyAI STT 3.7 Realtime Agent vs Google Chirp 3.

Last updated Methodology by Dileep Chagam

Who is better at what

STT 3.7 Realtime AgentChirp 2
  • Overall WER2.41%3.01%
  • Pipecat Dataset WER1.68%2.59%
  • Ocular Dataset WER3.81%3.80%
  • Time to first text493ms4.50s
  • Final-text delay47ms568ms

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 AgentChirp 2
  • ELO rating10241001–1039 · rank 1–16979966–999 · rank 9–21
  • Expected win rate56%44%
  • Ocular turns WER · 160 draws of 19727 won10 won
  • Pipecat WER · 834 draws of 999127 won38 won
  • Semantic WER · 178 draws of 19712 won7 won
  • Phone line · 159 draws of 19829 won10 won
  • Call-center noise · 114 draws of 19758 won25 won
  • Babble noise · 100 draws of 19763 won34 won
  • TV noise · 114 draws of 19835 won49 won
  • Reverb · 149 draws of 19842 won7 won
  • Short replies · 58 draws of 7010 won2 won
  • Entities · 44 draws of 495 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 Chirp 2 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 Google Chirp 2 3.01%, so AssemblyAI STT 3.7 Realtime Agent makes fewer errors by 0.59 points. On the 300 licensed Ocular turns, AssemblyAI STT 3.7 Realtime Agent scores 1.68% and Google Chirp 2 2.59%, so AssemblyAI STT 3.7 Realtime Agent makes fewer errors by 0.91 points. Word error rate counts incorrect, missing, and extra words, so lower is better.

  • Yes. AssemblyAI STT 3.7 Realtime Agent finishes the transcript 4.00s sooner. AssemblyAI STT 3.7 Realtime Agent has the final transcript 493ms after the speaker stops and Google Chirp 2 4.50s, 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 4.00s 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 Google Chirp 2 after 3.80%, at the median.

Test your voice agent the same way.

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