Fish Audio Transcribe Stream 1 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)
  • Transcribe Stream 11.90% · 123ms
  • Finch-Global3.17% · 51ms

Fish Audio Transcribe Stream 1 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

Transcribe Stream 1Finch-Global
  • Overall WER2.21%3.33%
  • Pipecat Dataset WER1.90%3.17%
  • Ocular Dataset WER2.78%3.62%
  • Time to first text730ms1.83s
  • Final-text delay123ms51ms

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.

Transcribe Stream 1Finch-Global
  • ELO rating10411035–1046 · rank 1–7998981–1017 · rank 5–21
  • Expected win rate56%44%
  • Ocular turns WER · 166 draws of 19821 won11 won
  • Pipecat WER · 813 draws of 999147 won39 won
  • Semantic WER · 190 draws of 1987 won1 won
  • Phone line · 161 draws of 19823 won14 won
  • Call-center noise · 122 draws of 19843 won33 won
  • Babble noise · 101 draws of 19852 won45 won
  • TV noise · 142 draws of 19632 won22 won
  • Reverb · 126 draws of 19761 won10 won
  • Short replies · 60 draws of 707 won3 won
  • Entities · 42 draws of 507 won1 won

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

Transcribe Stream 1 vs Finch-Global FAQ

  • Fish Audio Transcribe Stream 1 is more accurate on both datasets. On the 1,000 public Pipecat clips, Fish Audio Transcribe Stream 1 scores 2.21% and Murf Finch-Global 3.33%, so Fish Audio Transcribe Stream 1 makes fewer errors by 1.12 points. On the 300 licensed Ocular turns, Fish Audio Transcribe Stream 1 scores 1.90% and Murf Finch-Global 3.17%, so Fish Audio Transcribe Stream 1 makes fewer errors by 1.27 points. Word error rate counts incorrect, missing, and extra words, so lower is better.

  • Yes. Fish Audio Transcribe Stream 1 finishes the transcript 1.10s sooner. Fish Audio Transcribe Stream 1 has the final transcript 730ms after the speaker stops and Murf Finch-Global 1.83s, at the median. Both are measured on the same 206 streamed turns.

  • Fish Audio Transcribe Stream 1 is the stronger pick for voice agents: it makes fewer errors on real conversations and finishes the transcript 1.10s 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. Fish Audio Transcribe Stream 1 shows first text after 2.78% 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.