ElevenLabs Scribe v3 Live vs Fish Audio Transcribe Stream 1
Accuracy and speed
Left gets fewer words wrong. Lower has the final text ready sooner.
- Scribe v3 Live1.81% · 57ms
- Transcribe Stream 11.90% · 123ms
ElevenLabs Scribe v3 Live is ahead of Fish Audio Transcribe Stream 1 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 WER1.89%2.21%
- Pipecat Dataset WER1.81%1.90%
- Ocular Dataset WER2.05%2.78%
- Time to first text685ms730ms
- Final-text delay57ms123ms
Lower is better on every measure. Too close to call means the likely ranges overlap.
Scribe v3 Live vs Transcribe Stream 1 FAQ
Which is more accurate, ElevenLabs Scribe v3 Live or Fish Audio Transcribe Stream 1?
ElevenLabs Scribe v3 Live is more accurate on the public clips. On the 1,000 public Pipecat clips, ElevenLabs Scribe v3 Live scores 1.89% and Fish Audio Transcribe Stream 1 2.21%, so ElevenLabs Scribe v3 Live makes fewer errors by 0.31 points. On the 300 licensed Ocular turns, ElevenLabs Scribe v3 Live scores 1.81% and Fish Audio Transcribe Stream 1 1.90%, with likely ranges of 1.59 to 2.04% and 1.68 to 2.14%. The ranges overlap, so the difference is too close to call. Word error rate counts incorrect, missing, and extra words, so lower is better.
ElevenLabs Scribe v3 Live is more accurate on the public clips. On the 1,000 public Pipecat clips, ElevenLabs Scribe v3 Live scores 1.89% and Fish Audio Transcribe Stream 1 2.21%, so ElevenLabs Scribe v3 Live makes fewer errors by 0.31 points. On the 300 licensed Ocular turns, ElevenLabs Scribe v3 Live scores 1.81% and Fish Audio Transcribe Stream 1 1.90%, with likely ranges of 1.59 to 2.04% and 1.68 to 2.14%. 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. ElevenLabs Scribe v3 Live finishes the transcript 45ms sooner. ElevenLabs Scribe v3 Live has the final transcript 685ms after the speaker stops and Fish Audio Transcribe Stream 1 730ms, at the median. Both are measured on the same 206 streamed turns.
ElevenLabs Scribe v3 Live finishes the transcript 45ms 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. ElevenLabs Scribe v3 Live shows first text after 2.05% and Fish Audio Transcribe Stream 1 after 2.78%, at the median.
Test your voice agent the same way.
Cekura runs these simulated calls on your agent and scores every one.