Reson8 Rumble-2 vs Speechmatics Linden
Accuracy and speed
Left gets fewer words wrong. Lower has the final text ready sooner.
- Rumble-21.69% · 18ms
- Linden2.46% · 147ms
Reson8 Rumble-2 is ahead of Speechmatics Linden 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 WER2.05%2.97%
- Pipecat Dataset WER1.69%2.46%
- Ocular Dataset WER2.75%3.93%
- Time to first text486ms1.19s
- Final-text delay18ms147ms
Lower is better on every measure. Too close to call means the likely ranges overlap.
Rumble-2 vs Linden FAQ
Which is more accurate, Reson8 Rumble-2 or Speechmatics Linden?
Reson8 Rumble-2 is more accurate on both datasets. On the 1,000 public Pipecat clips, Reson8 Rumble-2 scores 2.05% and Speechmatics Linden 2.97%, so Reson8 Rumble-2 makes fewer errors by 0.91 points. On the 300 licensed Ocular turns, Reson8 Rumble-2 scores 1.69% and Speechmatics Linden 2.46%, so Reson8 Rumble-2 makes fewer errors by 0.77 points. Word error rate counts incorrect, missing, and extra words, so lower is better.
Reson8 Rumble-2 is more accurate on both datasets. On the 1,000 public Pipecat clips, Reson8 Rumble-2 scores 2.05% and Speechmatics Linden 2.97%, so Reson8 Rumble-2 makes fewer errors by 0.91 points. On the 300 licensed Ocular turns, Reson8 Rumble-2 scores 1.69% and Speechmatics Linden 2.46%, so Reson8 Rumble-2 makes fewer errors by 0.77 points. Word error rate counts incorrect, missing, and extra words, so lower is better.
Yes. Reson8 Rumble-2 finishes the transcript 704ms sooner. Reson8 Rumble-2 has the final transcript 486ms after the speaker stops and Speechmatics Linden 1.19s, at the median. Both are measured on the same 206 streamed turns.
Reson8 Rumble-2 is the stronger pick for voice agents: it makes fewer errors on real conversations and finishes the transcript 704ms 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. Reson8 Rumble-2 shows first text after 2.75% and Speechmatics Linden after 3.93%, at the median.
Test your voice agent the same way.
Cekura runs these simulated calls on your agent and scores every one.