NVIDIA Nemotron 3.5 Transcribe vs Reson8 Rumble-2
Accuracy and speed
Left gets fewer words wrong. Lower has the final text ready sooner. Nemotron 3.5 Transcribe doesn't support streaming, so it is not shown.
- Nemotron 3.5 Transcribe1.89%
- Rumble-21.69% · 18ms
Reson8 Rumble-2 is ahead of NVIDIA Nemotron 3.5 Transcribe on both datasets together. The Pipecat Dataset and The Ocular Dataset are too close to call.
Last updated Methodology by Dileep Chagam
Who is better at what
- Overall WER2.28%2.05%
- Pipecat Dataset WER1.89%1.69%
- Ocular Dataset WER3.03%2.75%
- Time to first textn/a486ms
- Final-text delayn/a18ms
Lower is better on every measure. Too close to call means the likely ranges overlap.
Nemotron 3.5 Transcribe vs Rumble-2 FAQ
Which is more accurate, NVIDIA Nemotron 3.5 Transcribe or Reson8 Rumble-2?
Reson8 Rumble-2 is more accurate on the public clips. On the 1,000 public Pipecat clips, NVIDIA Nemotron 3.5 Transcribe scores 2.28% and Reson8 Rumble-2 2.05%, so Reson8 Rumble-2 makes fewer errors by 0.23 points. On the 300 licensed Ocular turns, NVIDIA Nemotron 3.5 Transcribe scores 1.89% and Reson8 Rumble-2 1.69%, with likely ranges of 1.63 to 2.17% and 1.48 to 1.92%. The ranges overlap, so the difference is too close to call. Word error rate counts incorrect, missing, and extra words, so lower is better.
Reson8 Rumble-2 is more accurate on the public clips. On the 1,000 public Pipecat clips, NVIDIA Nemotron 3.5 Transcribe scores 2.28% and Reson8 Rumble-2 2.05%, so Reson8 Rumble-2 makes fewer errors by 0.23 points. On the 300 licensed Ocular turns, NVIDIA Nemotron 3.5 Transcribe scores 1.89% and Reson8 Rumble-2 1.69%, with likely ranges of 1.63 to 2.17% and 1.48 to 1.92%. The ranges overlap, so the difference is too close to call. Word error rate counts incorrect, missing, and extra words, so lower is better.
They cannot be compared on speed. NVIDIA Nemotron 3.5 Transcribe was tested on whole recordings rather than a live stream, so its speed is not measured.
The two are too close to call on real conversations. NVIDIA Nemotron 3.5 Transcribe was tested on whole recordings rather than a live stream, so its speed is not measured. A voice agent cannot reply until the final transcript arrives, and the Ocular recordings are real conversations.
NVIDIA Nemotron 3.5 Transcribe was tested on whole recordings rather than a live stream, so whether it streams partial text was not measured. Reson8 Rumble-2 returns partial text before the speaker stops.
Test your voice agent the same way.
Cekura runs these simulated calls on your agent and scores every one.