Voice AI BenchmarksBy

SPEECHMATICSSPEECH-TO-TEXT146,760 SAMPLES / 30 DAYSLAST RUN AUG 28, 2026, 21:30 UTC

official resource

Enhanced speech-to-text benchmarks

Enhanced is Speechmatics' higher-accuracy real-time transcription tier.

Time to Final Segment#17 / 24
341ms
Word Error Rate#4 / 28
4.3%
Time to First Token#12 / 24
1536ms

Overview

Speechmatics exposes Enhanced as a distinct real-time operating point on a service that also supports multilingual transcription, diarization and code switching.

Enhanced 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
Speechmatics
Hosted by
Speechmatics
Source
Official API
Licensing
Proprietary
Deployment
On-prem
Region
Europe
Features
Multilingual, VAD, Diarization, Translation, Code switching, Keyterm biasing

How Enhanced 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 Enhanced highlighted.
  1. #1STT RT v564 ms
  2. #2Parakeet TDT 0.6B v370 ms
  3. #3STT 183 ms
  4. #4Nova 399 ms
  5. #5Nova 2101 ms
  6. #6Ink 2108 ms
  7. #7Scribe v2 Realtime120 ms
  8. #8Universal 3.5 Pro146 ms
  9. #9Defaultvia Azure155 ms
  10. #10Whisper Large v3180 ms
  11. #11Velma 2 STT Streaming191 ms
  12. #12Grok STT199 ms
  13. #13Pulse205 ms
  14. #14Defaultvia Speechmatics223 ms
  15. #15Defaultvia Gradium249 ms
  16. #16resonant-1288 ms
  17. #17Enhanced341 ms
Show all 24 models
  1. #18Voxtral Mini Transcribe Realtime 2602356 ms
  2. #19GPT Realtime Whisper559 ms
  3. #20GPT-4o mini Transcribe623 ms
  4. #21Solaria 1680 ms
  5. #22GPT-4o Transcribe742 ms
  6. #23Chirp 2811 ms
  7. #24Chirp 3813 ms
median of all models · 202 ms
STT models over the last 30 days, ranked on Time to Final Segment.
#ModelHostTTFSWERTTFTSamples
1STT RT v5Soniox64 ms1533 ms29,322
2Parakeet TDT 0.6B v3Together AI70 ms1210 ms29,161
3STT 1Inworld AI83 ms1468 ms29,280
4Nova 3Deepgram99 ms1434 ms29,290
5Nova 2Deepgram101 ms1436 ms29,145
6Ink 2Cartesia108 ms1812 ms29,288
7Scribe v2 RealtimeElevenLabs120 ms2157 ms29,313
8Universal 3.5 ProAssemblyAI146 ms1030 ms29,280
9DefaultAzure155 ms1792 ms6,682
10Whisper Large v3Together AI180 ms1241 ms29,171
11Velma 2 STT StreamingModulate191 ms1576 ms17,895
12Grok STTxAI199 ms—29,322
13PulseSmallest205 ms2072 ms29,313
14DefaultSpeechmatics223 ms1473 ms29,323
15DefaultGradium249 ms1979 ms28,235
16resonant-1Reson8288 ms—17,066
17EnhancedSpeechmatics341 ms1536 ms29,322
18Voxtral Mini Transcribe Realtime 2602Mistral356 ms1828 ms29,175
19GPT Realtime WhisperOpenAI559 ms1823 ms29,286
20GPT-4o mini TranscribeOpenAI623 ms—29,318
21Solaria 1Gladia680 ms1703 ms26,356
22GPT-4o TranscribeOpenAI742 ms—29,318
23Chirp 2Google811 ms6000 ms29,237
24Chirp 3Google813 ms5999 ms29,319
—FluxDeepgram—1089 ms29,354
—Flux MultilingualDeepgram—1175 ms29,344
—Nemotron 3.5 ASR StreamingTogether AI—1540 ms29,193
—Universal StreamingAssemblyAI—1510 ms29,297
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: 4th of 28 on Word Error Rate.

Latency vs accuracy

TTFS vs WEREach point is one measured model · Enhanced highlighted · 30-day averagesTime to Final Segment against Word Error Rate for every measured STT model, with Enhanced highlighted.
STT RT v5 — 64 ms, 5.8%Parakeet TDT 0.6B v3 — 70 ms, 11.0%STT 1 — 83 ms, 4.8%Nova 3 — 99 ms, 6.3%Nova 2 — 101 ms, 7.9%Ink 2 — 108 ms, 4.9%Scribe v2 Realtime — 120 ms, 5.5%Universal 3.5 Pro — 146 ms, 3.2%Default via Azure — 155 ms, 5.3%Whisper Large v3 — 180 ms, 8.2%Velma 2 STT Streaming — 191 ms, 6.4%Grok STT — 199 ms, 4.8%Pulse — 205 ms, 5.3%Default via Speechmatics — 223 ms, 5.5%Default via Gradium — 249 ms, 10.0%resonant-1 — 288 ms, 3.4%Voxtral Mini Transcribe Realtime 2602 — 356 ms, 6.2%GPT Realtime Whisper — 559 ms, 5.2%GPT-4o mini Transcribe — 623 ms, 5.1%Solaria 1 — 680 ms, 7.1%GPT-4o Transcribe — 742 ms, 4.7%Chirp 2 — 811 ms, 5.0%Chirp 3 — 813 ms, 4.1%Enhanced — 341 ms, 4.3%

Where the errors come from

WER compositionEnhanced's Word Error Rate split by error type · 30-day averageEnhanced's WER split into substitutions, deletions and insertions.
  • Enhanced4.3%
SubstitutionsDeletionsInsertions

Averages and tail latency

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

Enhanced latency distributionMilliseconds · shared axis across metrics · last 30 daysEnhanced's latency percentiles per metric: p25–p75 band, p50 tick, whisker to p99.
  • Time to Final Segmentp50 321 ms · p99 648 ms
  • Time to First Tokenp50 1565 ms · p99 3063 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 Segment341 ms281 ms321 ms378 ms448 ms502 ms648 ms29,322
Word Error Rate4.3%0.0%0.0%6.3%12.5%18.5%33.3%29,356
Time to First Token1536 ms1175 ms1565 ms1653 ms2032 ms2297 ms3063 ms29,356

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 daysEnhanced's daily median Time to Final Segment over the last 30 days.
Enhanced · Jul 30: 415 msEnhanced · Jul 31: 356 msEnhanced · Aug 1: 310 msEnhanced · Aug 2: 294 msEnhanced · Aug 3: 316 msEnhanced · Aug 4: 330 msEnhanced · Aug 5: 366 msEnhanced · Aug 6: 397 msEnhanced · Aug 7: 354 msEnhanced · Aug 8: 308 msEnhanced · Aug 9: 293 msEnhanced · Aug 10: 363 msEnhanced · Aug 11: 347 msEnhanced · Aug 12: 333 msEnhanced · Aug 13: 343 msEnhanced · Aug 14: 343 msEnhanced · Aug 15: 275 msEnhanced · Aug 16: 281 msEnhanced · Aug 17: 307 msEnhanced · Aug 18: 325 msEnhanced · Aug 19: 328 msEnhanced · Aug 20: 341 msEnhanced · Aug 21: 329 msEnhanced · Aug 22: 298 msEnhanced · Aug 23: 301 msEnhanced · Aug 24: 352 msEnhanced · Aug 25: 309 msEnhanced · Aug 26: 310 msEnhanced · Aug 27: 281 msEnhanced · Aug 28: 315 ms
Word Error Rate — daily averageDaily average · UTC daysEnhanced's daily Word Error Rate over the last 30 days.
Enhanced · Jul 30: 6.0%Enhanced · Jul 31: 6.6%Enhanced · Aug 1: 4.4%Enhanced · Aug 2: 5.2%Enhanced · Aug 3: 4.1%Enhanced · Aug 4: 5.7%Enhanced · Aug 5: 6.0%Enhanced · Aug 6: 5.5%Enhanced · Aug 7: 6.0%Enhanced · Aug 8: 4.7%Enhanced · Aug 9: 5.5%Enhanced · Aug 10: 5.1%Enhanced · Aug 11: 4.7%Enhanced · Aug 12: 5.0%Enhanced · Aug 13: 4.7%Enhanced · Aug 14: 5.5%Enhanced · Aug 15: 5.1%Enhanced · Aug 16: 5.8%Enhanced · Aug 17: 4.8%Enhanced · Aug 18: 6.5%Enhanced · Aug 19: 5.4%Enhanced · Aug 20: 5.8%Enhanced · Aug 21: 5.8%Enhanced · Aug 22: 6.6%Enhanced · Aug 23: 4.0%Enhanced · Aug 24: 4.9%Enhanced · Aug 25: 5.0%Enhanced · Aug 26: 4.8%Enhanced · Aug 27: 5.8%Enhanced · Aug 28: 5.8%

Time to Final Segment by dataset

Enhanced by test conditionMilliseconds · lower is better · best condition firstEnhanced's Time to Final Segment on each benchmark dataset.
  1. PipeCat (production)339 ms
  2. WildASR phone codec340 ms
  3. WildASR clean341 ms
  4. WildASR far-field342 ms
  5. WildASR reverb342 ms
  6. WildASR noise gaps343 ms
  7. WildASR accents347 ms
  8. WildASR clipping349 ms
Time to Final Segment per dataset, with the number of samples behind each figure.
DatasetTTFSSamples
LibriSpeech371 ms1
ProductionPipeCat339 ms14,939
AccentsWildASR347 ms1,438
CleanWildASR341 ms5,779
ClippingWildASR349 ms1,443
Far-fieldWildASR342 ms1,442
Noise gapsWildASR343 ms1,441
Phone codecWildASR340 ms1,441
ReverbWildASR342 ms1,398

Strongest condition: PipeCat (production) at 339 ms · weakest: LibriSpeech at 371 ms.

Limits of this comparison

Enhanced remains separate from the provider's default tier so the extra recognition mode is not blended into a company-level average.

  • The benchmark measures transcription outcomes and timing, not translation quality or every language available through Speechmatics.

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