Deepgram Nova-3 vs GPT-4o Transcribe
Deepgram Nova-3 is ahead of GPT-4o Transcribe on the Ocular Dataset and final-text delay. The Pipecat Dataset is too close to call.
Last updated Methodology by Dileep Chagam
Accuracy vs speed
Each dot is one model. Bottom-left is best: fewer errors and a faster final transcript.
Which to pick
Deepgram Nova-3 is the better pick over GPT-4o Transcribe on every measure that separates them: accuracy on real phone conversations (8.06 points lower) and quick turn-taking once the caller stops (549ms sooner). For accuracy on clean, scripted audio, the gap is inside the margin of error, so neither is ahead.
The four metrics, side by side
- Deepgram Nova-3
- 3.46%12th
- GPT-4o Transcribe
- 3.90%14th
Too close to call: inside the margin of error
- Deepgram Nova-3
- 4.39%11th
- GPT-4o Transcribe
- 12.45%17th
Deepgram Nova-3, 8.06 points lower
- Deepgram Nova-3
- 1.06s7th
- GPT-4o Transcribe
- No text until the speaker stops
Not comparable
- Deepgram Nova-3
- 103ms4th
- GPT-4o Transcribe
- 652ms17th
Deepgram Nova-3, 549ms sooner
Lower is better on every metric. The small figure is each model's place in the whole field. What each metric measures.
- WER · Word error rate
- Share of words the model got wrong: replaced, missing or added. Lower is better.
- Pipecat Dataset
- 1,000 short public clips from Pipecat's STT benchmark dataset.
- Ocular Dataset
- 8 recordings from 4 real phone conversations, licensed from Ocular.
- TTFT · Time to first text
- How long after the audio starts streaming until the first words appear.
- TTFS · Final-text delay
- How long after the caller stops until the full transcript arrives. A voice agent waits for this before it replies.
First text and final text, at three percentiles
Deepgram Nova-3 shows first text after 2.07s and final text after 155ms, and GPT-4o Transcribe shows first text after 40.63s and final text after 1.97s, on a bad turn (P95) across the same streamed turns.
| Measure | Model | P50 | P90 | P95 |
|---|---|---|---|---|
| Time to first text | Deepgram Nova-3 | 1.06s | 1.10s | 2.07s |
| GPT-4o Transcribe | No text until the speaker stops | |||
| Final-text delay | Deepgram Nova-3 | 103ms | 143ms | 155ms |
| GPT-4o Transcribe | 652ms | 1.73s | 1.97s | |
- P50 / P90 / P95 · Typical, slow and slowest turns
- Half of turns are faster than P50. Only 1 in 10 is slower than P90, and 1 in 20 slower than P95.
How each ranks among all 17 models
Place among the 17 models on each metric. Lower is better on all four.
| Metric | Nova-3 | 4o Transcribe |
|---|---|---|
| Pipecat Dataset WER | 3.46%12th of 17 | 3.90%14th of 17 |
| Ocular Dataset WER | 4.39%11th of 17 | 12.45%17th of 17 |
| Time to first text | 1.06s7th of 15 | Not comparable |
| Final-text delay | 103ms4th of 17 | 652ms17th of 17 |
How much audio each model was tested on
| Model | Pipecat clips | Ocular recordings | Turns timed | Failed |
|---|---|---|---|---|
| Deepgram Nova-3 | 1,000 | 8 | 202 | 0 |
| GPT-4o Transcribe | 1,000 | 8 | 206 | 0 |
Only clips with usable text count, so the two totals can differ. How WER is scored.
Related comparisons
Both providers, every model: Deepgram vs OpenAI
- Deepgram Nova-3 vs GPT-4o Mini Transcribe
- Deepgram Nova-3 vs Smallest Pulse
- Deepgram Nova-3 vs Deepgram Flux English
- Deepgram Nova-3 vs Inworld STT-1
- Deepgram Nova-3 vs Soniox STT-RT v5
- Deepgram Nova-3 vs Speechmatics Linden
- Deepgram Nova-3 vs Cartesia Ink 2
- Deepgram Nova-3 vs GPT Realtime Whisper
- Deepgram Nova-3 vs Reson8
- Deepgram Nova-3 vs Google Chirp 3
- Deepgram Nova-3 vs Deepgram Flux Multilingual
- Deepgram Nova-3 vs AssemblyAI Universal 3.6 Pro
- Deepgram Nova-3 vs Gradium
- GPT-4o Transcribe vs Deepgram Flux English
- GPT-4o Transcribe vs Smallest Pulse
- GPT-4o Transcribe vs GPT-4o Mini Transcribe
- GPT-4o Transcribe vs Deepgram Flux Multilingual
- GPT-4o Transcribe vs Inworld STT-1
- GPT-4o Transcribe vs Soniox STT-RT v5
- GPT-4o Transcribe vs Speechmatics Linden
- GPT-4o Transcribe vs Cartesia Ink 2
- GPT-4o Transcribe vs GPT Realtime Whisper
- GPT-4o Transcribe vs Reson8
- GPT-4o Transcribe vs Google Chirp 3
- GPT-4o Transcribe vs AssemblyAI Universal 3.6 Pro
- GPT-4o Transcribe vs Gradium
Frequently asked questions
Which is more accurate, Deepgram Nova-3 or GPT-4o Transcribe?
Deepgram Nova-3 is more accurate on real conversations. On the 1,000 public Pipecat clips, Deepgram Nova-3 scores 3.46% and GPT-4o Transcribe 3.90%, with likely ranges of 3.16 to 3.78% and 3.43 to 4.42%. The ranges overlap, so the difference is too close to call. On the eight licensed Ocular recordings, Deepgram Nova-3 scores 4.39% and GPT-4o Transcribe 12.45%, so Deepgram Nova-3 makes fewer errors by 8.06 points. Word error rate counts incorrect, missing, and extra words, so lower is better.
Is Deepgram Nova-3 faster than GPT-4o Transcribe?
Yes. Deepgram Nova-3 finishes the transcript 549ms sooner. Deepgram Nova-3 has the final transcript 103ms after the speaker stops and GPT-4o Transcribe 652ms, at the median. Both are measured on the same 206 streamed turns.
Is Deepgram Nova-3 or GPT-4o Transcribe better for voice agents?
Deepgram Nova-3 is the stronger pick for voice agents: it makes fewer errors on real conversations and finishes the transcript 549ms sooner. A voice agent cannot reply until the final transcript arrives, and the Ocular recordings are real conversations.
Do Deepgram Nova-3 and GPT-4o Transcribe stream text while the speaker is talking?
Only Deepgram Nova-3. GPT-4o Transcribe returns no text until the speaker stops, which matters if your agent reacts to partial transcripts.