Left gets fewer words wrong. Lower has the final text ready sooner.
Smallest Pulse is ahead on the Ocular Dataset and time to first text, and Zoom Scribe on the Pipecat Dataset and final-text delay.
Last updated Methodology by Dileep Chagam
Lower is better on every measure. Too close to call means the likely ranges overlap.
Zoom Scribe is more accurate on clean public clips and Smallest Pulse on real conversations. On the 1,000 public Pipecat clips, Smallest Pulse scores 3.66% and Zoom Scribe 1.97%, so Zoom Scribe makes fewer errors by 1.69 points. On the eight licensed Ocular recordings, Smallest Pulse scores 3.46% and Zoom Scribe 4.88%, so Smallest Pulse makes fewer errors by 1.43 points. Word error rate counts incorrect, missing, and extra words, so lower is better.
Zoom Scribe is more accurate on clean public clips and Smallest Pulse on real conversations. On the 1,000 public Pipecat clips, Smallest Pulse scores 3.66% and Zoom Scribe 1.97%, so Zoom Scribe makes fewer errors by 1.69 points. On the eight licensed Ocular recordings, Smallest Pulse scores 3.46% and Zoom Scribe 4.88%, so Smallest Pulse makes fewer errors by 1.43 points. Word error rate counts incorrect, missing, and extra words, so lower is better.
No. Zoom Scribe finishes the transcript 83ms sooner. Smallest Pulse has the final transcript 201ms after the speaker stops and Zoom Scribe 118ms, at the median. Both are measured on the same 206 streamed turns.
It is a trade-off. Smallest Pulse makes fewer errors on real conversations, while Zoom Scribe finishes the transcript 83ms 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. Smallest Pulse shows first text after 1.06s and Zoom Scribe after 1.67s, at the median, so Smallest Pulse is 611ms sooner.