Deepgram Flux English vs GPT-4o Transcribe
Accuracy and speed
Left gets fewer words wrong. Lower has the final text ready sooner.
- Flux English3.90% · 108ms
- 4o Transcribe3.90% · 652ms
Deepgram Flux English is ahead of GPT-4o Transcribe on both datasets together, the Ocular Dataset and final-text delay. The two tie on the Pipecat Dataset.
Last updated Methodology by Dileep Chagam
Who is better at what
- Overall WER4.00%6.85%
- Pipecat Dataset WER3.90%3.90%
- Ocular Dataset WER4.18%12.45%
- Time to first text791msn/a
- Final-text delay108ms652ms
Lower is better on every measure. Too close to call means the likely ranges overlap. For costs, see our breakdown of Deepgram pricing.
Flux English vs 4o Transcribe FAQ
Which is more accurate, Deepgram Flux English or GPT-4o Transcribe?
Deepgram Flux English is more accurate on the public clips. On the 1,000 public Pipecat clips, Deepgram Flux English scores 4.00% and GPT-4o Transcribe 6.85%, so Deepgram Flux English makes fewer errors by 2.85 points. On the eight licensed Ocular recordings, Deepgram Flux English scores 3.90% and GPT-4o Transcribe 3.90%, which is a tie. Word error rate counts incorrect, missing, and extra words, so lower is better.
Deepgram Flux English is more accurate on the public clips. On the 1,000 public Pipecat clips, Deepgram Flux English scores 4.00% and GPT-4o Transcribe 6.85%, so Deepgram Flux English makes fewer errors by 2.85 points. On the eight licensed Ocular recordings, Deepgram Flux English scores 3.90% and GPT-4o Transcribe 3.90%, which is a tie. Word error rate counts incorrect, missing, and extra words, so lower is better.
Final-text delay was not measured for one or both, so speed cannot be compared. Both are measured on the same 206 streamed turns.
Deepgram Flux English and GPT-4o Transcribe are level on the two figures a voice agent depends on most. A voice agent cannot reply until the final transcript arrives, and the Ocular recordings are real conversations.
Only Deepgram Flux English. GPT-4o Transcribe returns no text until the speaker stops, which matters if your agent reacts to partial transcripts.
Test your voice agent the same way.
Cekura runs these simulated calls on your agent and scores every one.