Deepgram Flux Multilingual vs GPT-4o Transcribe
Deepgram Flux Multilingual is ahead on the Ocular Dataset and final-text delay, and GPT-4o Transcribe on the Pipecat Dataset.
Quality and speed frontier
Every model's error rate on the public clips, against how long its transcript takes to settle.
The four metrics, side by side
Word error rate on 1,000 clips from Pipecat's STT benchmark dataset.
- Deepgram Flux Multilingual
- 4.96%14th
- GPT-4o Transcribe
- 3.90%12th
GPT-4o Transcribe, 1.06 points lower
Word error rate on eight licensed recordings provided by Ocular.
- Deepgram Flux Multilingual
- 4.53%13th
- GPT-4o Transcribe
- 12.45%15th
Deepgram Flux Multilingual, 7.92 points lower
Median delay before the first words arrive, from the first audio packet.
- Deepgram Flux Multilingual
- 793ms3rd
- GPT-4o Transcribe
- No text until the speaker stops
Not comparable
Median delay after the speaker stops before the transcript is final.
- Deepgram Flux Multilingual
- 114ms4th
- GPT-4o Transcribe
- 652ms15th
Deepgram Flux Multilingual, 538ms sooner
Lower is better on every metric. The small figure is each model's place in the whole field.
First text and final text, at three percentiles
Both measured on the same 206 streamed Ocular Dataset turns. P90 and P95 say what a caller meets on a bad turn, which a median hides.
| Measure | Model | P50 | P90 | P95 |
|---|---|---|---|---|
| Time to first text | Deepgram Flux Multilingual | 793ms | 2.00s | 3.41s |
| GPT-4o Transcribe | No text until the speaker stops | |||
| Final-text delay | Deepgram Flux Multilingual | 114ms | 135ms | 141ms |
| GPT-4o Transcribe | 652ms | 1.73s | 1.97s | |
Both against every other model
Open on the metric where the two sit furthest apart, with the other three a tab away.
Final-text delay after the speaker stops. Lower is better.
| # | Model | Final text | turns | |
|---|---|---|---|---|
| 1 | Inworld STT-1 | 41ms | 206 | |
| 2 | Deepgram Nova-3 | 103ms | 202 | |
| 3 | Deepgram Flux English | 108ms | 199 | |
| 4 | Deepgram Flux Multilingual | 114ms | 175 | |
| 5 | Cartesia Ink 2 | 123ms | 198 | |
| 6 | Speechmatics Linden | 147ms | 194 | |
| 7 | AssemblyAI Universal 3.5 Pro | 180ms | 205 | |
| 8 | Smallest Pulse | 201ms | 206 | |
| 9 | Gradium | 212ms | 198 | |
| 10 | Reson8 | 235ms | 201 | |
| 11 | Google Chirp 3 | 475ms | 206 | |
| 12 | GPT Realtime Whisper | 543ms | 206 | |
| 13 | Google Chirp 2 | 568ms | 205 | |
| 14 | GPT-4o Mini Transcribe | 630ms | 206 | |
| 15 | GPT-4o Transcribe | 652ms | 206 |
Median across the same 206 Ocular Dataset turns. Providers finalize differently, so compare models with similar finalization contracts before treating small gaps as meaningful.
Both on the other two views
Pipecat vs Ocular accuracy
Every model's error rate on the public clips, against the same model on real conversations.
Transcription response time
How soon each model returns its first words, against how soon its transcript is final.
What the numbers rest on
| Model | Pipecat clips | Ocular recordings | Turns timed | Failed |
|---|---|---|---|---|
| Deepgram Flux Multilingual | 999 | 8 | 175 | 31 |
| GPT-4o Transcribe | 1,000 | 8 | 206 | 0 |
A clip counts only where the model returned usable text, so counts differ between models. Word error rate is measured the same way throughout: incorrect, missing, and extra words against the reference transcript.
Frequently asked questions
Is Deepgram Flux Multilingual or GPT-4o Transcribe more accurate?
On the 1,000 public Pipecat clips, Deepgram Flux Multilingual scores 4.96% and GPT-4o Transcribe 3.90%, so GPT-4o Transcribe makes fewer errors by 1.06 points. On the eight licensed Ocular recordings, Deepgram Flux Multilingual scores 4.53% and GPT-4o Transcribe 12.45%, so Deepgram Flux Multilingual makes fewer errors by 7.92 points. The two datasets disagree, so the better choice depends on whether your audio looks more like clean public clips or real conversations.
Which is faster, Deepgram Flux Multilingual or GPT-4o Transcribe?
GPT-4o Transcribe returns no text while the speaker is talking, so time to first text cannot be compared. Deepgram Flux Multilingual has the final transcript 538ms sooner, 114ms after the speaker stops against 652ms. Both are measured on the same 206 streamed turns.
How were Deepgram Flux Multilingual and GPT-4o Transcribe compared?
Both ran the same two datasets under the same conditions as every other model in the benchmark: 1,000 public clips from Pipecat's STT benchmark dataset and eight licensed conversation recordings from Ocular. Deepgram Flux Multilingual returned usable text for 999 of the public clips and GPT-4o Transcribe for 1,000. Latency uses each provider's own supported finalization contract.