Murf Finch-Global vs GPT Realtime Whisper

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
  • Realtime Whisper2.16% · 543ms

GPT Realtime Whisper is ahead on both datasets together, the Pipecat Dataset and time to first text, and Murf Finch-Global on final-text delay. The Ocular Dataset is too close to call.

Last updated Methodology by Dileep Chagam

Who is better at what

Finch-GlobalRealtime Whisper
  • Overall WER3.33%2.63%
  • Pipecat Dataset WER3.17%2.16%
  • Ocular Dataset WER3.62%3.53%
  • Time to first text1.83s1.38s
  • Final-text delay51ms543ms

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-GlobalRealtime Whisper
  • ELO rating998981–1017 · rank 5–2110091005–1020 · rank 5–16
  • Expected win rate48%52%
  • Ocular turns WER · 158 draws of 19827 won13 won
  • Pipecat WER · 809 draws of 99954 won136 won
  • Semantic WER · 186 draws of 1985 won7 won
  • Phone line · 155 draws of 19826 won17 won
  • Call-center noise · 120 draws of 19847 won31 won
  • Babble noise · 112 draws of 19842 won44 won
  • TV noise · 102 draws of 1989 won87 won
  • Reverb · 122 draws of 19736 won39 won
  • Short replies · 56 draws of 7011 won3 won
  • Entities · 42 draws of 502 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 Realtime Whisper FAQ

  • GPT Realtime Whisper is more accurate on both datasets. On the 1,000 public Pipecat clips, Murf Finch-Global scores 3.33% and GPT Realtime Whisper 2.63%, so GPT Realtime Whisper makes fewer errors by 0.70 points. On the 300 licensed Ocular turns, Murf Finch-Global scores 3.17% and GPT Realtime Whisper 2.16%, so GPT Realtime Whisper makes fewer errors by 1.01 points. Word error rate counts incorrect, missing, and extra words, so lower is better.

  • No. GPT Realtime Whisper finishes the transcript 454ms sooner. Murf Finch-Global has the final transcript 1.83s after the speaker stops and GPT Realtime Whisper 1.38s, at the median. Both are measured on the same 206 streamed turns.

  • GPT Realtime Whisper is the stronger pick for voice agents: it makes fewer errors on real conversations and finishes the transcript 454ms 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. Murf Finch-Global shows first text after 3.62% and GPT Realtime Whisper after 3.53%, at the median.

Test your voice agent the same way.

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