Deepgram Nova-3vsZoom Scribe Pro
Accuracy and speed
Left gets fewer words wrong. Lower has the final text ready sooner.
- Nova-33.46% · 103ms
- Scribe Pro1.96% · 162ms
Deepgram Nova-3 is ahead on time to first text and final-text delay, and Zoom Scribe Pro on the Pipecat Dataset. The Ocular Dataset is too close to call.
Last updated Methodology by Dileep Chagam
Who is better at what
- Pipecat Dataset WER3.46%1.96%
- Ocular Dataset WER4.39%4.84%
- Time to first text1.06s1.70s
- Final-text delay103ms162ms
Lower is better on every measure. Too close to call means the likely ranges overlap.
Related comparisons
Both providers, every model: Deepgram vs Zoom
Nova-3 vs Scribe Pro FAQ
Which is more accurate, Deepgram Nova-3 or Zoom Scribe Pro?
Zoom Scribe Pro is more accurate on the public clips. On the 1,000 public Pipecat clips, Deepgram Nova-3 scores 3.46% and Zoom Scribe Pro 1.96%, so Zoom Scribe Pro makes fewer errors by 1.50 points. On the eight licensed Ocular recordings, Deepgram Nova-3 scores 4.39% and Zoom Scribe Pro 4.84%, with likely ranges of 3.42 to 5.25% and 4.23 to 5.43%. The ranges overlap, so the difference is too close to call. Word error rate counts incorrect, missing, and extra words, so lower is better.
Zoom Scribe Pro is more accurate on the public clips. On the 1,000 public Pipecat clips, Deepgram Nova-3 scores 3.46% and Zoom Scribe Pro 1.96%, so Zoom Scribe Pro makes fewer errors by 1.50 points. On the eight licensed Ocular recordings, Deepgram Nova-3 scores 4.39% and Zoom Scribe Pro 4.84%, with likely ranges of 3.42 to 5.25% and 4.23 to 5.43%. 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. Deepgram Nova-3 finishes the transcript 59ms sooner. Deepgram Nova-3 has the final transcript 103ms after the speaker stops and Zoom Scribe Pro 162ms, at the median. Both are measured on the same 206 streamed turns.
Deepgram Nova-3 finishes the transcript 59ms 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. Deepgram Nova-3 shows first text after 1.06s and Zoom Scribe Pro after 1.70s, at the median, so Deepgram Nova-3 is 635ms sooner.