Gradium vs Inworld STT-1
Accuracy and speed
Left gets fewer words wrong. Lower has the final text ready sooner.
- Gradium6.48% · 212ms
- STT-12.72% · 41ms
Inworld STT-1 is ahead of Gradium on both datasets together, the Pipecat Dataset, time to first text and final-text delay. The Ocular Dataset is too close to call.
Last updated Methodology by Dileep Chagam
Who is better at what
- Overall WER5.78%5.44%
- Pipecat Dataset WER6.48%2.72%
- Ocular Dataset WER4.45%10.60%
- Time to first text1.71s1.06s
- Final-text delay212ms41ms
Lower is better on every measure. Too close to call means the likely ranges overlap. Our guide explains how word error rate is measured.
Gradium vs STT-1 FAQ
Which is more accurate, Gradium or Inworld STT-1?
Inworld STT-1 is more accurate on both datasets. On the 1,000 public Pipecat clips, Gradium scores 5.78% and Inworld STT-1 5.44%, so Inworld STT-1 makes fewer errors by 0.34 points. On the eight licensed Ocular recordings, Gradium scores 6.48% and Inworld STT-1 2.72%, so Inworld STT-1 makes fewer errors by 3.76 points. Word error rate counts incorrect, missing, and extra words, so lower is better.
Inworld STT-1 is more accurate on both datasets. On the 1,000 public Pipecat clips, Gradium scores 5.78% and Inworld STT-1 5.44%, so Inworld STT-1 makes fewer errors by 0.34 points. On the eight licensed Ocular recordings, Gradium scores 6.48% and Inworld STT-1 2.72%, so Inworld STT-1 makes fewer errors by 3.76 points. Word error rate counts incorrect, missing, and extra words, so lower is better.
No. Inworld STT-1 finishes the transcript 642ms sooner. Gradium has the final transcript 1.71s after the speaker stops and Inworld STT-1 1.06s, at the median. Both are measured on the same 206 streamed turns.
Inworld STT-1 is the stronger pick for voice agents: it makes fewer errors on real conversations and finishes the transcript 642ms 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. Gradium shows first text after 4.45% and Inworld STT-1 after 10.60%, at the median.
Test your voice agent the same way.
Cekura runs these simulated calls on your agent and scores every one.