Deepgram Nova-3 vs Murf Finch-Global

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)
  • Nova-33.46% · 103ms
  • Finch-Global3.17% · 51ms

Murf Finch-Global is ahead on both datasets together and final-text delay, and Deepgram Nova-3 on time to first text. The Pipecat Dataset and The Ocular Dataset are too close to call.

Last updated Methodology by Dileep Chagam

Who is better at what

Nova-3Finch-Global
  • Overall WER3.78%3.33%
  • Pipecat Dataset WER3.46%3.17%
  • Ocular Dataset WER4.39%3.62%
  • Time to first text1.06s1.83s
  • Final-text delay103ms51ms

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.

Nova-3Finch-Global
  • ELO rating971940–1015 · rank 5–22998981–1017 · rank 5–21
  • Expected win rate46%54%
  • Ocular turns WER · 153 draws of 19817 won28 won
  • Pipecat WER · 781 draws of 999103 won115 won
  • Semantic WER · 186 draws of 1987 won5 won
  • Phone line · 151 draws of 19818 won29 won
  • Call-center noise · 107 draws of 19724 won66 won
  • Babble noise · 92 draws of 19822 won84 won
  • TV noise · 122 draws of 19821 won55 won
  • Reverb · 132 draws of 19741 won24 won
  • Short replies · 51 draws of 705 won14 won
  • Entities · 40 draws of 507 won3 won

Short replies are decided on the error count and entities on exact capture, with no draw margin. How the rating is worked out

Nova-3 vs Finch-Global FAQ

  • Murf Finch-Global is more accurate on the public clips. On the 1,000 public Pipecat clips, Deepgram Nova-3 scores 3.78% and Murf Finch-Global 3.33%, so Murf Finch-Global makes fewer errors by 0.46 points. On the 300 licensed Ocular turns, Deepgram Nova-3 scores 3.46% and Murf Finch-Global 3.17%, with likely ranges of 3.16 to 3.78% and 2.84 to 3.51%. 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 769ms sooner. Deepgram Nova-3 has the final transcript 1.06s after the speaker stops and Murf Finch-Global 1.83s, at the median. Both are measured on the same 206 streamed turns.

  • Deepgram Nova-3 finishes the transcript 769ms 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 4.39% and Murf Finch-Global after 3.62%, at the median.

Test your voice agent the same way.

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