AssemblyAI Universal 3.5 Pro vs Smallest Pulse
AssemblyAI Universal 3.5 Pro is ahead of Smallest Pulse on the Pipecat Dataset, time to first text and final-text delay. The Ocular Dataset is too close to call.
AssemblyAI Universal 3.5 Pro has a newer release. See AssemblyAI Universal 3.6 Pro vs Smallest Pulse.
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 Smallest Pulse on every measure that separates them: accuracy on clean, scripted audio (1.73 points lower), showing words while the caller is still talking (568ms sooner) and quick turn-taking once the caller stops (21ms sooner). For 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
- Smallest Pulse
- 3.66%13th
AssemblyAI Universal 3.5 Pro, 1.73 points lower
- AssemblyAI Universal 3.5 Pro
- 4.23%10th
- Smallest Pulse
- 3.46%3rd
Too close to call: inside the margin of error
- AssemblyAI Universal 3.5 Pro
- 489ms2nd
- Smallest Pulse
- 1.06s5th
AssemblyAI Universal 3.5 Pro, 568ms sooner
- AssemblyAI Universal 3.5 Pro
- 180ms9th
- Smallest Pulse
- 201ms10th
AssemblyAI Universal 3.5 Pro, 21ms 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 Smallest Pulse shows first text after 3.03s and final text after 773ms, 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 |
| Smallest Pulse | 1.06s | 2.25s | 3.03s | |
| Final-text delay | AssemblyAI Universal 3.5 Pro | 180ms | 532ms | 692ms |
| Smallest Pulse | 201ms | 356ms | 773ms |
- 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.
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 | Pulse |
|---|---|---|
| Pipecat Dataset WER | 1.93%2nd of 17 | 3.66%13th of 17 |
| Ocular Dataset WER | 4.23%10th of 17 | 3.46%3rd of 17 |
| Time to first text | 489ms2nd of 15 | 1.06s5th of 15 |
| Final-text delay | 180ms9th of 17 | 201ms10th 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 |
| Smallest Pulse | 1,000 | 8 | 206 | 0 |
Only clips with usable text count, so the two totals can differ. How WER is scored.
Related comparisons
- AssemblyAI Universal 3.5 Pro vs AssemblyAI Universal 3.6 Pro
- Smallest Pulse vs Deepgram Nova-3
- Smallest Pulse vs GPT-4o Transcribe
- Smallest Pulse vs Deepgram Flux English
- Smallest Pulse vs GPT-4o Mini Transcribe
- Smallest Pulse vs Inworld STT-1
- Smallest Pulse vs Soniox STT-RT v5
- Smallest Pulse vs Speechmatics Linden
- Smallest Pulse vs Cartesia Ink 2
- Smallest Pulse vs Deepgram Flux Multilingual
- Smallest Pulse vs GPT Realtime Whisper
- Smallest Pulse vs Reson8
- Smallest Pulse vs Google Chirp 3
- Smallest Pulse vs AssemblyAI Universal 3.6 Pro
- Smallest Pulse vs Gradium
Frequently asked questions
Which is more accurate, AssemblyAI Universal 3.5 Pro or Smallest Pulse?
AssemblyAI Universal 3.5 Pro is more accurate on the public clips. On the 1,000 public Pipecat clips, AssemblyAI Universal 3.5 Pro scores 1.93% and Smallest Pulse 3.66%, so AssemblyAI Universal 3.5 Pro makes fewer errors by 1.73 points. On the eight licensed Ocular recordings, AssemblyAI Universal 3.5 Pro scores 4.23% and Smallest Pulse 3.46%, with likely ranges of 3.57 to 4.68% and 2.66 to 4.22%. 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 Smallest Pulse?
Yes. AssemblyAI Universal 3.5 Pro finishes the transcript 21ms sooner. AssemblyAI Universal 3.5 Pro has the final transcript 180ms after the speaker stops and Smallest Pulse 201ms, at the median. Both are measured on the same 206 streamed turns.
Is AssemblyAI Universal 3.5 Pro or Smallest Pulse better for voice agents?
AssemblyAI Universal 3.5 Pro finishes the transcript 21ms 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 Smallest Pulse 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 Smallest Pulse after 1.06s, at the median, so AssemblyAI Universal 3.5 Pro is 568ms sooner.