Voice AI BenchmarksBy

OPENAISPEECH-TO-TEXT91,435 SAMPLES / 30 DAYSLAST RUN SEP 15, 2026, 08:30 UTC

official resource

GPT-4o Transcribe speech-to-text benchmarks

GPT-4o Transcribe, hosted by OpenAI, measures mean 754 ms time to final segment (25th of 28) and 4.6% word error rate (9th of 30) among STT systems. Results cover the last 30 days. Coval benchmarks hosted endpoints daily using a consistent evaluation methodology. Last measured .

GPT-4o Transcribe is OpenAI's GPT-4o-based speech recognition model for the Audio API.

Time to Final Segment#25 / 28
754ms
Word Error Rate#9 / 30
4.6%

Overview

GPT-4o Transcribe is a dedicated transcription model rather than the transcription path inside an OpenAI Realtime session.

GPT-4o Transcribe is tested every day on fixed public audio — clean, accented, noisy, reverberant, far-field, clipped and phone-codec speech — for transcription accuracy and streaming latency.

Technical specifications

Made by
OpenAI
Hosted by
OpenAI
Source
Official API
Licensing
Proprietary
Deployment
Cloud
Region
US
Features
Keyterm biasing, Multilingual, VAD

How GPT-4o Transcribe ranks

Full STT dashboard
Time to Final Segment — every measured STT modelMilliseconds · lower is better · 30-day averageEvery STT model ranked on Time to Final Segment, with GPT-4o Transcribe highlighted.
  1. #1Qwen3 ASR 1.7b39 ms
  2. #2Qwen3 ASR Fast46 ms
  3. #3STT RT v557 ms
  4. #4STT 165 ms
  5. #5Parakeet TDT 0.6B v381 ms
  6. #6Nova 389 ms
  7. #7Nova 292 ms
  8. #8Flux Multilingual98 ms
  9. #9Flux99 ms
  10. #10Ink 2122 ms
  11. #11Whisper Large v3via Baseten125 ms
  12. #12Scribe v2 Realtime133 ms
  13. #13Universal 3.5 Pro173 ms
  14. #14Grok STT207 ms
  15. #15Defaultvia Speechmatics209 ms
  16. #16Pulse212 ms
  17. #17Defaultvia Gradium246 ms
  18. #18resonant-1264 ms
  19. #19Whisper Large v3via Together AI295 ms
  20. #20Enhanced299 ms
  21. #21Gemini 3.5 Transcribe Live305 ms
  22. #22Voxtral Mini Transcribe Realtime 2602401 ms
  23. #23GPT Realtime Whisper551 ms
  24. #24GPT-4o mini Transcribe690 ms
  25. #25GPT-4o Transcribe754 ms
Show all 28 models
  1. #26Chirp 3776 ms
  2. #27Solaria 1805 ms
  3. #28Chirp 2873 ms
median of all models · 208 ms
STT models over the last 30 days, ranked on Time to Final Segment.
#ModelHostTTFSWERTTFTSamples
1Qwen3 ASR 1.7bBaseten39 ms931 ms1,256
2Qwen3 ASR FastNari46 ms1748 ms4,451
3STT RT v5Soniox57 ms1529 ms22,468
4STT 1Inworld AI65 ms1400 ms22,836
5Parakeet TDT 0.6B v3Together AI81 ms1215 ms22,488
6Nova 3Deepgram89 ms1419 ms22,855
7Nova 2Deepgram92 ms1420 ms22,733
8Flux MultilingualDeepgram98 ms1164 ms12,251
9FluxDeepgram99 ms1076 ms12,243
10Ink 2Cartesia122 ms1827 ms22,862
11Whisper Large v3Baseten125 ms912 ms1,252
12Scribe v2 RealtimeElevenLabs133 ms2175 ms22,866
13Universal 3.5 ProAssemblyAI173 ms1041 ms20,253
14Grok STTxAI207 ms—22,853
15DefaultSpeechmatics209 ms1428 ms22,876
16PulseSmallest212 ms2011 ms22,856
17DefaultGradium246 ms1980 ms22,839
18resonant-1Reson8264 ms—22,860
19Whisper Large v3Together AI295 ms1338 ms22,744
20EnhancedSpeechmatics299 ms1492 ms22,876
21Gemini 3.5 Transcribe LiveGemini305 ms1679 ms16,403
22Voxtral Mini Transcribe Realtime 2602Mistral401 ms1851 ms22,583
23GPT Realtime WhisperOpenAI551 ms1814 ms22,812
24GPT-4o mini TranscribeOpenAI690 ms—22,830
25GPT-4o TranscribeOpenAI754 ms—22,831
26Chirp 3Google776 ms5974 ms22,875
27Solaria 1Gladia805 ms1804 ms22,439
28Chirp 2Google873 ms6071 ms22,789
—Nemotron 3.5 ASR StreamingTogether AI—1548 ms22,747
—Universal StreamingAssemblyAI—1514 ms20,235
Under-sampled models are excluded; tied models share a place. Dotted WER values split into substitutions, deletions and insertions on hover or tap.

Highest relative placement: 9th of 30 on Word Error Rate.

Latency vs accuracy

TTFS vs WEREach point is one measured model · GPT-4o Transcribe highlighted · 30-day averagesTime to Final Segment against Word Error Rate for every measured STT model, with GPT-4o Transcribe highlighted.
Qwen3 ASR 1.7b — 39 ms, 4.1%Qwen3 ASR Fast — 46 ms, 3.2%STT RT v5 — 57 ms, 5.7%STT 1 — 65 ms, 4.4%Parakeet TDT 0.6B v3 — 81 ms, 11.2%Nova 3 — 89 ms, 6.2%Nova 2 — 92 ms, 7.9%Flux Multilingual — 98 ms, 8.8%Flux — 99 ms, 6.7%Ink 2 — 122 ms, 4.8%Whisper Large v3 via Baseten — 125 ms, 5.6%Scribe v2 Realtime — 133 ms, 5.5%Universal 3.5 Pro — 173 ms, 3.1%Grok STT — 207 ms, 4.6%Default via Speechmatics — 209 ms, 5.4%Pulse — 212 ms, 5.3%Default via Gradium — 246 ms, 10.0%resonant-1 — 264 ms, 3.4%Whisper Large v3 via Together AI — 295 ms, 8.5%Enhanced — 299 ms, 4.3%Gemini 3.5 Transcribe Live — 305 ms, 3.9%Voxtral Mini Transcribe Realtime 2602 — 401 ms, 6.2%GPT Realtime Whisper — 551 ms, 5.1%GPT-4o mini Transcribe — 690 ms, 4.9%Chirp 3 — 776 ms, 4.1%Solaria 1 — 805 ms, 7.1%Chirp 2 — 873 ms, 4.9%GPT-4o Transcribe — 754 ms, 4.6%

Where the errors come from

WER compositionGPT-4o Transcribe's Word Error Rate split by error type · 30-day averageGPT-4o Transcribe's WER split into substitutions, deletions and insertions.
  • GPT-4o Transcribe4.6%
SubstitutionsDeletionsInsertions

Averages and tail latency

Averages hide slow outliers — these are the distributions behind each figure.

GPT-4o Transcribe latency distributionMilliseconds · shared axis across metrics · last 30 daysGPT-4o Transcribe's latency percentiles per metric: p25–p75 band, p50 tick, whisker to p99.
  • Time to Final Segmentp50 725 ms · p99 1374 ms
band p25–p75 · tick p50 · whisker to p99 with p90 and p95 stops
Average and percentile values per metric, with the number of samples behind each row.
MetricAveragep25p50p75p90p95p99Samples
Time to Final Segment754 ms628 ms725 ms826 ms936 ms1035 ms1374 ms22,831
Word Error Rate4.6%0.0%0.0%5.3%12.5%21.1%53.3%22,868

Last 30 days

Daily medians from the same measurement runs · gaps are days without qualifying runs.

Time to Final Segment — daily p50Line p50 · band p25–p75 · UTC daysGPT-4o Transcribe's daily median Time to Final Segment over the last 30 days.
GPT-4o Transcribe · Sep 1: 674 msGPT-4o Transcribe · Sep 2: 734 msGPT-4o Transcribe · Sep 3: 748 msGPT-4o Transcribe · Sep 4: 980 msGPT-4o Transcribe · Sep 5: 708 msGPT-4o Transcribe · Sep 6: 689 msGPT-4o Transcribe · Sep 7: 699 msGPT-4o Transcribe · Sep 8: 752 msGPT-4o Transcribe · Sep 9: 747 msGPT-4o Transcribe · Sep 10: 754 msGPT-4o Transcribe · Sep 11: 739 msGPT-4o Transcribe · Sep 12: 655 msGPT-4o Transcribe · Sep 13: 709 msGPT-4o Transcribe · Sep 14: 750 msGPT-4o Transcribe · Sep 15: 747 ms
Word Error Rate — daily averageDaily average · UTC daysGPT-4o Transcribe's daily Word Error Rate over the last 30 days.
GPT-4o Transcribe · Sep 1: 4.3%GPT-4o Transcribe · Sep 2: 4.8%GPT-4o Transcribe · Sep 3: 5.6%GPT-4o Transcribe · Sep 4: 5.4%GPT-4o Transcribe · Sep 5: 5.3%GPT-4o Transcribe · Sep 6: 5.1%GPT-4o Transcribe · Sep 7: 5.3%GPT-4o Transcribe · Sep 8: 5.4%GPT-4o Transcribe · Sep 9: 5.4%GPT-4o Transcribe · Sep 10: 5.6%GPT-4o Transcribe · Sep 11: 4.9%GPT-4o Transcribe · Sep 12: 5.3%GPT-4o Transcribe · Sep 13: 5.1%GPT-4o Transcribe · Sep 14: 5.1%GPT-4o Transcribe · Sep 15: 7.0%

Time to Final Segment by dataset

GPT-4o Transcribe by test conditionMilliseconds · lower is better · best condition firstGPT-4o Transcribe's Time to Final Segment on each benchmark dataset.
  1. WildASR accents652 ms
  2. WildASR reverb740 ms
  3. WildASR phone codec758 ms
  4. PipeCat (production)759 ms
  5. WildASR clipping762 ms
  6. WildASR clean762 ms
  7. WildASR far-field764 ms
  8. WildASR noise gaps768 ms
Time to Final Segment per dataset, with the number of samples behind each figure.
DatasetTTFSSamples
LibriSpeech1028 ms1
ProductionPipeCat759 ms11,496
AccentsWildASR652 ms1,115
CleanWildASR762 ms4,569
ClippingWildASR762 ms1,138
Far-fieldWildASR764 ms1,141
Noise gapsWildASR768 ms1,141
Phone codecWildASR758 ms1,136
ReverbWildASR740 ms1,094

Strongest condition: WildASR accents at 652 ms · weakest: LibriSpeech at 1028 ms.

How fast is GPT-4o Transcribe?

On OpenAI, GPT-4o Transcribe measures mean 754 ms time to final segment (25th of 28). Last measured 2026-09-15.

How accurate is GPT-4o Transcribe?

On OpenAI, GPT-4o Transcribe measures 4.6% word error rate (9th of 30). Last measured 2026-09-15.

Who hosts GPT-4o Transcribe?

GPT-4o Transcribe is created by OpenAI and served by OpenAI. Coval measures each hosted endpoint separately.

Limits of this comparison

Coval reports accuracy, first partial output and final-segment timing separately because a fast partial does not guarantee a fast settled transcript.

  • Prompting, language mix and audio format can affect production behavior beyond Coval's fixed public datasets.

Official sources

Results are re-measured daily using fixed datasets and reported over a rolling 30-day window. Same datasets, prompts and metric definitions for every model, measured by Coval’s open-source runner. Full methodology on the overview.

Evaluate your own voice agent

Use Coval to test your production configuration, prompts and calls—not only the public benchmark endpoints.

Book a Demo