Voice AI BenchmarksBy

INWORLD AISPEECH-TO-TEXT146,564 SAMPLES / 30 DAYSLAST RUN AUG 28, 2026, 21:30 UTC

official resource

STT 1 speech-to-text benchmarks

Inworld STT 1 is Inworld's streaming transcription model for interactive systems.

Time to Final Segment#3 / 24
83ms
Word Error Rate#7 / 28
4.8%
Time to First Token#8 / 24
1468ms

Overview

The tested endpoint is multilingual and includes voice activity plus keyterm biasing.

STT 1 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
Inworld AI
Hosted by
Inworld AI
Source
Official API
Licensing
Proprietary
Deployment
Cloud
Region
US
Features
Multilingual, VAD, Keyterm biasing

How STT 1 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 STT 1 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
Show all 24 models
  1. #13Pulse205 ms
  2. #14Defaultvia Speechmatics223 ms
  3. #15Defaultvia Gradium249 ms
  4. #16resonant-1288 ms
  5. #17Enhanced341 ms
  6. #18Voxtral Mini Transcribe Realtime 2602356 ms
  7. #19GPT Realtime Whisper559 ms
  8. #20GPT-4o mini Transcribe623 ms
  9. #21Solaria 1680 ms
  10. #22GPT-4o Transcribe742 ms
  11. #23Chirp 2811 ms
  12. #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.

Latency vs accuracy

TTFS vs WEREach point is one measured model · STT 1 highlighted · 30-day averagesTime to Final Segment against Word Error Rate for every measured STT model, with STT 1 highlighted.
STT RT v5 — 64 ms, 5.8%Parakeet TDT 0.6B v3 — 70 ms, 11.0%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%Enhanced — 341 ms, 4.3%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%STT 1 — 83 ms, 4.8%

Where the errors come from

WER compositionSTT 1's Word Error Rate split by error type · 30-day averageSTT 1's WER split into substitutions, deletions and insertions.
  • STT 14.8%
SubstitutionsDeletionsInsertions

Averages and tail latency

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

STT 1 latency distributionMilliseconds · shared axis across metrics · last 30 daysSTT 1's latency percentiles per metric: p25–p75 band, p50 tick, whisker to p99.
  • Time to Final Segmentp50 72 ms · p99 148 ms
  • Time to First Tokenp50 1489 ms · p99 2635 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 Segment83 ms60 ms72 ms106 ms115 ms120 ms148 ms29,280
Word Error Rate4.8%0.0%0.0%6.5%12.5%19.0%50.0%29,321
Time to First Token1468 ms1087 ms1489 ms1780 ms1984 ms2090 ms2635 ms29,321

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 daysSTT 1's daily median Time to Final Segment over the last 30 days.
STT 1 · Jul 30: 112 msSTT 1 · Jul 31: 104 msSTT 1 · Aug 1: 110 msSTT 1 · Aug 2: 103 msSTT 1 · Aug 3: 112 msSTT 1 · Aug 4: 107 msSTT 1 · Aug 5: 108 msSTT 1 · Aug 6: 107 msSTT 1 · Aug 7: 107 msSTT 1 · Aug 8: 107 msSTT 1 · Aug 9: 106 msSTT 1 · Aug 10: 111 msSTT 1 · Aug 11: 103 msSTT 1 · Aug 12: 63 msSTT 1 · Aug 13: 60 msSTT 1 · Aug 14: 60 msSTT 1 · Aug 15: 60 msSTT 1 · Aug 16: 60 msSTT 1 · Aug 17: 62 msSTT 1 · Aug 18: 62 msSTT 1 · Aug 19: 63 msSTT 1 · Aug 20: 66 msSTT 1 · Aug 21: 62 msSTT 1 · Aug 22: 60 msSTT 1 · Aug 23: 67 msSTT 1 · Aug 24: 63 msSTT 1 · Aug 25: 67 msSTT 1 · Aug 26: 59 msSTT 1 · Aug 27: 61 msSTT 1 · Aug 28: 64 ms
Word Error Rate — daily averageDaily average · UTC daysSTT 1's daily Word Error Rate over the last 30 days.
STT 1 · Jul 30: 3.5%STT 1 · Jul 31: 5.5%STT 1 · Aug 1: 9.6%STT 1 · Aug 2: 7.5%STT 1 · Aug 3: 5.1%STT 1 · Aug 4: 8.2%STT 1 · Aug 5: 5.7%STT 1 · Aug 6: 4.8%STT 1 · Aug 7: 6.0%STT 1 · Aug 8: 6.8%STT 1 · Aug 9: 8.7%STT 1 · Aug 10: 4.5%STT 1 · Aug 11: 6.5%STT 1 · Aug 12: 7.7%STT 1 · Aug 13: 8.1%STT 1 · Aug 14: 6.1%STT 1 · Aug 15: 5.0%STT 1 · Aug 16: 6.5%STT 1 · Aug 17: 6.8%STT 1 · Aug 18: 5.8%STT 1 · Aug 19: 5.9%STT 1 · Aug 20: 8.3%STT 1 · Aug 21: 5.5%STT 1 · Aug 22: 7.1%STT 1 · Aug 23: 8.4%STT 1 · Aug 24: 5.5%STT 1 · Aug 25: 5.6%STT 1 · Aug 26: 5.1%STT 1 · Aug 27: 4.6%STT 1 · Aug 28: 5.8%

Time to Final Segment by dataset

STT 1 by test conditionMilliseconds · lower is better · best condition firstSTT 1's Time to Final Segment on each benchmark dataset.
  1. WildASR phone codec67 ms
  2. WildASR accents79 ms
  3. WildASR reverb81 ms
  4. PipeCat (production)81 ms
  5. WildASR clean87 ms
  6. WildASR clipping87 ms
  7. WildASR noise gaps88 ms
  8. WildASR far-field91 ms
Time to Final Segment per dataset, with the number of samples behind each figure.
DatasetTTFSSamples
LibriSpeech81 ms1
ProductionPipeCat81 ms14,932
AccentsWildASR79 ms1,438
CleanWildASR87 ms5,753
ClippingWildASR87 ms1,444
Far-fieldWildASR91 ms1,441
Noise gapsWildASR88 ms1,439
Phone codecWildASR67 ms1,441
ReverbWildASR81 ms1,391

Strongest condition: WildASR phone codec at 67 ms · weakest: WildASR far-field at 91 ms.

Limits of this comparison

Coval measures it independently from Inworld's TTS 2 family so recognition and synthesis remain attributable to the correct stage.

  • Coval does not evaluate game-character behavior or other application features in Inworld's broader platform.

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