Deepgram Flux Multilingual 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)
  • Flux Multilingual4.96% · 114ms
  • Finch-Global3.17% · 51ms

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

Last updated Methodology by Dileep Chagam

Who is better at what

Flux MultilingualFinch-Global
  • Overall WER4.81%3.33%
  • Pipecat Dataset WER4.96%3.17%
  • Ocular Dataset WER4.53%3.62%
  • Time to first text793ms1.83s
  • Final-text delay114ms51ms

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.

Flux MultilingualFinch-Global
  • ELO rating936916–966 · rank 18–23998981–1017 · rank 5–21
  • Expected win rate41%59%
  • Ocular turns WER · 146 draws of 19616 won34 won
  • Pipecat WER · 746 draws of 99969 won184 won
  • Semantic WER · 180 draws of 1966 won10 won
  • Phone line · 147 draws of 19817 won34 won
  • Call-center noise · 96 draws of 19517 won82 won
  • Babble noise · 82 draws of 18524 won79 won
  • TV noise · 114 draws of 19517 won64 won
  • Reverb · 112 draws of 19739 won46 won
  • Short replies · 49 draws of 696 won14 won
  • Entities · 38 draws of 495 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

Flux Multilingual vs Finch-Global FAQ

  • Murf Finch-Global is more accurate on both datasets. On the 1,000 public Pipecat clips, Deepgram Flux Multilingual scores 4.81% and Murf Finch-Global 3.33%, so Murf Finch-Global makes fewer errors by 1.49 points. On the 300 licensed Ocular turns, Deepgram Flux Multilingual scores 4.96% and Murf Finch-Global 3.17%, so Murf Finch-Global makes fewer errors by 1.79 points. Word error rate counts incorrect, missing, and extra words, so lower is better.

  • Yes. Deepgram Flux Multilingual finishes the transcript 1.04s sooner. Deepgram Flux Multilingual has the final transcript 793ms after the speaker stops and Murf Finch-Global 1.83s, at the median. Both are measured on the same 206 streamed turns.

  • It is a trade-off. Murf Finch-Global makes fewer errors on real conversations, while Deepgram Flux Multilingual finishes the transcript 1.04s 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. Deepgram Flux Multilingual shows first text after 4.53% 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.