GPT Realtime Whisper vs Reson8
Accuracy and speed
Left gets fewer words wrong. Lower has the final text ready sooner.
- Realtime Whisper2.16% · 543ms
- Reson82.10% · 235ms
Reson8 is ahead on both datasets together and final-text delay, and GPT Realtime Whisper 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
- Overall WER2.63%2.38%
- Pipecat Dataset WER2.16%2.10%
- Ocular Dataset WER3.53%2.93%
- Time to first text1.38s2.40s
- Final-text delay543ms235ms
Lower is better on every measure. Too close to call means the likely ranges overlap. Our guide covers testing OpenAI Realtime API voice agents.
Realtime Whisper vs Reson8 FAQ
Which is more accurate, GPT Realtime Whisper or Reson8?
Reson8 is more accurate on the public clips. On the 1,000 public Pipecat clips, GPT Realtime Whisper scores 2.63% and Reson8 2.38%, so Reson8 makes fewer errors by 0.24 points. On the eight licensed Ocular recordings, GPT Realtime Whisper scores 2.16% and Reson8 2.10%, with likely ranges of 1.92 to 2.41% and 1.86 to 2.35%. 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 is more accurate on the public clips. On the 1,000 public Pipecat clips, GPT Realtime Whisper scores 2.63% and Reson8 2.38%, so Reson8 makes fewer errors by 0.24 points. On the eight licensed Ocular recordings, GPT Realtime Whisper scores 2.16% and Reson8 2.10%, with likely ranges of 1.92 to 2.41% and 1.86 to 2.35%. 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. GPT Realtime Whisper finishes the transcript 1.02s sooner. GPT Realtime Whisper has the final transcript 1.38s after the speaker stops and Reson8 2.40s, at the median. Both are measured on the same 206 streamed turns.
GPT Realtime Whisper finishes the transcript 1.02s 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. GPT Realtime Whisper shows first text after 3.53% and Reson8 after 2.93%, at the median.
Test your voice agent the same way.
Cekura runs these simulated calls on your agent and scores every one.