AssemblyAI Universal 3.5 Pro vs Google Chirp 3
AssemblyAI Universal 3.5 Pro is ahead of Google Chirp 3 on time to first text and final-text delay. The Pipecat Dataset and The Ocular Dataset are too close to call.
AssemblyAI Universal 3.5 Pro has a newer release. See AssemblyAI Universal 3.6 Pro vs Google Chirp 3.
Last updated Methodology by Dileep Chagam
Accuracy vs speed
Each dot is one model. Bottom-left is best: fewer errors and a faster final transcript.
Which to pick
AssemblyAI Universal 3.5 Pro is the better pick over Google Chirp 3 on every measure that separates them: showing words while the caller is still talking (3.99s sooner) and quick turn-taking once the caller stops (295ms sooner). For accuracy on clean, scripted audio and accuracy on real phone conversations, the gap is inside the margin of error, so neither is ahead.
The four metrics, side by side
- AssemblyAI Universal 3.5 Pro
- 1.93%2nd
- Google Chirp 3
- 2.00%3rd
Too close to call: inside the margin of error
- AssemblyAI Universal 3.5 Pro
- 4.23%10th
- Google Chirp 3
- 4.19%9th
Too close to call: inside the margin of error
- AssemblyAI Universal 3.5 Pro
- 489ms2nd
- Google Chirp 3
- 4.48s14th
AssemblyAI Universal 3.5 Pro, 3.99s sooner
- AssemblyAI Universal 3.5 Pro
- 180ms9th
- Google Chirp 3
- 475ms13th
AssemblyAI Universal 3.5 Pro, 295ms sooner
Lower is better on every metric. The small figure is each model's place in the whole field. What each metric measures.
- WER · Word error rate
- Share of words the model got wrong: replaced, missing or added. Lower is better.
- Pipecat Dataset
- 1,000 short public clips from Pipecat's STT benchmark dataset.
- Ocular Dataset
- 8 recordings from 4 real phone conversations, licensed from Ocular.
- TTFT · Time to first text
- How long after the audio starts streaming until the first words appear.
- TTFS · Final-text delay
- How long after the caller stops until the full transcript arrives. A voice agent waits for this before it replies.
First text and final text, at three percentiles
AssemblyAI Universal 3.5 Pro shows first text after 1.08s and final text after 692ms, and Google Chirp 3 shows first text after 7.97s and final text after 1.41s, on a bad turn (P95) across the same streamed turns.
| Measure | Model | P50 | P90 | P95 |
|---|---|---|---|---|
| Time to first text | AssemblyAI Universal 3.5 Pro | 489ms | 966ms | 1.08s |
| Google Chirp 3 | 4.48s | 6.21s | 7.97s | |
| Final-text delay | AssemblyAI Universal 3.5 Pro | 180ms | 532ms | 692ms |
| Google Chirp 3 | 475ms* | 941ms* | 1.41s* |
- P50 / P90 / P95 · Typical, slow and slowest turns
- Half of turns are faster than P50. Only 1 in 10 is slower than P90, and 1 in 20 slower than P95.
* Final-text delay is the observed time after stream close, because controlled finalization is not available for this model. Compare it with care against models that finalize on request.
How each ranks among all 17 models
Place among the 17 models on each metric. Lower is better on all four.
| Metric | Universal 3.5 Pro | Chirp 3 |
|---|---|---|
| Pipecat Dataset WER | 1.93%2nd of 17 | 2.00%3rd of 17 |
| Ocular Dataset WER | 4.23%10th of 17 | 4.19%9th of 17 |
| Time to first text | 489ms2nd of 15 | 4.48s14th of 15 |
| Final-text delay | 180ms9th of 17 | 475ms13th of 17 |
How much audio each model was tested on
| Model | Pipecat clips | Ocular recordings | Turns timed | Failed |
|---|---|---|---|---|
| AssemblyAI Universal 3.5 Pro | 1,000 | 8 | 205 | 1 |
| Google Chirp 3 | 1,000 | 8 | 206 | 0 |
Only clips with usable text count, so the two totals can differ. How WER is scored.
Related comparisons
Both providers, every model: AssemblyAI vs Google
- AssemblyAI Universal 3.5 Pro vs AssemblyAI Universal 3.6 Pro
- Google Chirp 3 vs Reson8
- Google Chirp 3 vs GPT Realtime Whisper
- Google Chirp 3 vs AssemblyAI Universal 3.6 Pro
- Google Chirp 3 vs Cartesia Ink 2
- Google Chirp 3 vs Speechmatics Linden
- Google Chirp 3 vs Soniox STT-RT v5
- Google Chirp 3 vs Google Chirp 2
- Google Chirp 3 vs Inworld STT-1
- Google Chirp 3 vs GPT-4o Mini Transcribe
- Google Chirp 3 vs Deepgram Nova-3
- Google Chirp 3 vs Smallest Pulse
- Google Chirp 3 vs GPT-4o Transcribe
- Google Chirp 3 vs Deepgram Flux English
- Google Chirp 3 vs Deepgram Flux Multilingual
- Google Chirp 3 vs Gradium
Frequently asked questions
Which is more accurate, AssemblyAI Universal 3.5 Pro or Google Chirp 3?
AssemblyAI Universal 3.5 Pro and Google Chirp 3 are too close to call on accuracy. On the 1,000 public Pipecat clips, AssemblyAI Universal 3.5 Pro scores 1.93% and Google Chirp 3 2.00%, with likely ranges of 1.69 to 2.18% and 1.77 to 2.25%. The ranges overlap, so the difference is too close to call. On the eight licensed Ocular recordings, AssemblyAI Universal 3.5 Pro scores 4.23% and Google Chirp 3 4.19%, with likely ranges of 3.57 to 4.68% and 2.97 to 5.27%. The ranges overlap, so the difference is too close to call. Word error rate counts incorrect, missing, and extra words, so lower is better.
Is AssemblyAI Universal 3.5 Pro faster than Google Chirp 3?
Yes. AssemblyAI Universal 3.5 Pro finishes the transcript 295ms sooner. AssemblyAI Universal 3.5 Pro has the final transcript 180ms after the speaker stops and Google Chirp 3 475ms, at the median. Both are measured on the same 206 streamed turns.
Is AssemblyAI Universal 3.5 Pro or Google Chirp 3 better for voice agents?
AssemblyAI Universal 3.5 Pro finishes the transcript 295ms 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.
Do AssemblyAI Universal 3.5 Pro and Google Chirp 3 stream text while the speaker is talking?
Yes, both return partial text before the speaker stops. AssemblyAI Universal 3.5 Pro shows first text after 489ms and Google Chirp 3 after 4.48s, at the median, so AssemblyAI Universal 3.5 Pro is 3.99s sooner.