Murf Finch-Global vs Smallest Pulse

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)
  • Finch-Global3.17% · 51ms
  • Pulse3.66% · 201ms

Murf Finch-Global is ahead on both datasets together and final-text delay, and Smallest Pulse 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

Finch-GlobalPulse
  • Overall WER3.33%3.59%
  • Pipecat Dataset WER3.17%3.66%
  • Ocular Dataset WER3.62%3.46%
  • Time to first text1.83s1.06s
  • Final-text delay51ms201ms

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.

Finch-GlobalPulse
  • ELO rating998981–1017 · rank 5–211005998–1011 · rank 6–18
  • Expected win rate49%51%
  • Ocular turns WER · 158 draws of 19830 won10 won
  • Pipecat WER · 790 draws of 999105 won104 won
  • Semantic WER · 181 draws of 1989 won8 won
  • Phone line · 150 draws of 19726 won21 won
  • Call-center noise · 101 draws of 19752 won44 won
  • Babble noise · 114 draws of 19833 won51 won
  • TV noise · 136 draws of 19826 won36 won
  • Reverb · 125 draws of 19720 won52 won
  • Short replies · 56 draws of 7013 won1 won
  • Entities · 41 draws of 503 won6 won

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

Finch-Global vs Pulse FAQ

  • Murf Finch-Global is more accurate on the public clips. On the 1,000 public Pipecat clips, Murf Finch-Global scores 3.33% and Smallest Pulse 3.59%, so Murf Finch-Global makes fewer errors by 0.26 points. On the 300 licensed Ocular turns, Murf Finch-Global scores 3.17% and Smallest Pulse 3.66%, with likely ranges of 2.84 to 3.51% and 3.34 to 3.99%. The ranges overlap, so the difference is too close to call. Word error rate counts incorrect, missing, and extra words, so lower is better.

  • No. Smallest Pulse finishes the transcript 776ms sooner. Murf Finch-Global has the final transcript 1.83s after the speaker stops and Smallest Pulse 1.06s, at the median. Both are measured on the same 206 streamed turns.

  • Smallest Pulse finishes the transcript 776ms 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. Murf Finch-Global shows first text after 3.62% and Smallest Pulse after 3.46%, at the median.

Test your voice agent the same way.

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