Inworld STT-1 vs Soniox STT-RT v5

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)
  • STT-12.72% · 41ms
  • STT-RT v52.54% · 38ms

Soniox STT-RT v5 is ahead on both datasets together and final-text delay, and Inworld STT-1 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

STT-1STT-RT v5
  • Overall WER5.44%3.02%
  • Pipecat Dataset WER2.72%2.54%
  • Ocular Dataset WER10.60%3.92%
  • Time to first text1.06s1.20s
  • Final-text delay41ms38ms

Lower is better on every measure. Too close to call means the likely ranges overlap. Our guide explains how word error rate is measured.

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.

STT-1STT-RT v5
  • ELO rating10311011–1043 · rank 1–131000986–1007 · rank 6–20
  • Expected win rate54%46%
  • Ocular turns WER · 156 draws of 19833 won9 won
  • Pipecat WER · 830 draws of 999105 won64 won
  • Semantic WER · 187 draws of 1989 won2 won
  • Phone line · 155 draws of 19732 won10 won
  • Call-center noise · 125 draws of 19847 won26 won
  • Babble noise · 111 draws of 19860 won27 won
  • TV noise · 119 draws of 19872 won7 won
  • Reverb · 147 draws of 19837 won14 won
  • Short replies · 49 draws of 7018 won3 won
  • Entities · 47 draws of 503 won0 won

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

STT-1 vs STT-RT v5 FAQ

  • Soniox STT-RT v5 is more accurate on the public clips. On the 1,000 public Pipecat clips, Inworld STT-1 scores 5.44% and Soniox STT-RT v5 3.02%, so Soniox STT-RT v5 makes fewer errors by 2.42 points. On the 300 licensed Ocular turns, Inworld STT-1 scores 2.72% and Soniox STT-RT v5 2.54%, with likely ranges of 1.67 to 4.68% and 2.27 to 2.83%. 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. Inworld STT-1 finishes the transcript 132ms sooner. Inworld STT-1 has the final transcript 1.06s after the speaker stops and Soniox STT-RT v5 1.20s, at the median. Both are measured on the same 206 streamed turns.

  • Inworld STT-1 finishes the transcript 132ms 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. Inworld STT-1 shows first text after 10.60% and Soniox STT-RT v5 after 3.92%, at the median.

Test your voice agent the same way.

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