Fish Audio Transcribe Stream 1 vs GPT Realtime Whisper
Accuracy and speed
Left gets fewer words wrong. Lower has the final text ready sooner.
- Transcribe Stream 11.90% · 123ms
- Realtime Whisper2.16% · 543ms
Fish Audio Transcribe Stream 1 is ahead of GPT Realtime Whisper on both datasets together, time to first text and final-text delay. The Pipecat Dataset and The Ocular Dataset are too close to call.
Last updated Methodology by Dileep Chagam
Who is better at what
- Overall WER2.21%2.63%
- Pipecat Dataset WER1.90%2.16%
- Ocular Dataset WER2.78%3.53%
- Time to first text730ms1.38s
- Final-text delay123ms543ms
Lower is better on every measure. Too close to call means the likely ranges overlap. Our guide covers testing OpenAI Realtime API voice agents.
Transcribe Stream 1 vs Realtime Whisper FAQ
Which is more accurate, Fish Audio Transcribe Stream 1 or GPT Realtime Whisper?
Fish Audio Transcribe Stream 1 is more accurate on the public clips. On the 1,000 public Pipecat clips, Fish Audio Transcribe Stream 1 scores 2.21% and GPT Realtime Whisper 2.63%, so Fish Audio Transcribe Stream 1 makes fewer errors by 0.42 points. On the eight licensed Ocular recordings, Fish Audio Transcribe Stream 1 scores 1.90% and GPT Realtime Whisper 2.16%, with likely ranges of 1.68 to 2.14% and 1.92 to 2.41%. The ranges overlap, so the difference is too close to call. Word error rate counts incorrect, missing, and extra words, so lower is better.
Fish Audio Transcribe Stream 1 is more accurate on the public clips. On the 1,000 public Pipecat clips, Fish Audio Transcribe Stream 1 scores 2.21% and GPT Realtime Whisper 2.63%, so Fish Audio Transcribe Stream 1 makes fewer errors by 0.42 points. On the eight licensed Ocular recordings, Fish Audio Transcribe Stream 1 scores 1.90% and GPT Realtime Whisper 2.16%, with likely ranges of 1.68 to 2.14% and 1.92 to 2.41%. 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. Fish Audio Transcribe Stream 1 finishes the transcript 649ms sooner. Fish Audio Transcribe Stream 1 has the final transcript 730ms after the speaker stops and GPT Realtime Whisper 1.38s, at the median. Both are measured on the same 206 streamed turns.
Fish Audio Transcribe Stream 1 finishes the transcript 649ms 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. Fish Audio Transcribe Stream 1 shows first text after 2.78% and GPT Realtime Whisper after 3.53%, at the median.
Test your voice agent the same way.
Cekura runs these simulated calls on your agent and scores every one.