Voice AI BenchmarksBy

GOOGLESPEECH-TO-TEXT132,924 SAMPLES / 30 DAYSLAST RUN OCT 16, 2026, 23:30 UTC

Gemini 3.5 Transcribe Live speech-to-text benchmarks

Gemini 3.5 Transcribe Live, hosted by Gemini, measures mean 323 ms time to final segment (21st of 29) and 3.0% word error rate (6th of 31) among STT systems. Results cover the last 30 days. Coval benchmarks hosted endpoints daily using a consistent evaluation methodology. Last measured .

Coval has active speech-to-text measurements for Gemini 3.5 Transcribe Live, created by Google and served on Gemini.

Time to Final Segment#21 / 29
323ms
Word Error Rate#6 / 31
3%
Time to First Token#16 / 27
1574ms

Overview

Gemini 3.5 Transcribe Live 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
Google
Hosted by
Gemini
Source
Official API
Licensing
Proprietary
Deployment
Cloud
Region
US
Features
Keyterm biasing, Multilingual, VAD

How Gemini 3.5 Transcribe Live 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 Gemini 3.5 Transcribe Live highlighted.
  1. #1Qwen3 ASR 1.7b39 ms
  2. #2STT RT v545 ms
  3. #3Qwen3 ASR Fast47 ms
  4. #4STT 168 ms
  5. #5Nova 385 ms
  6. #6Nova 287 ms
  7. #7Parakeet TDT 0.6B v396 ms
  8. #8Flux Multilingual98 ms
  9. #9Flux99 ms
  10. #10Scribe v2 Realtime120 ms
  11. #11Whisper Large v3via Baseten132 ms
  12. #12Universal 3.6 Pro134 ms
  13. #13Ink 2149 ms
  14. #14Universal 3.5 Pro165 ms
  15. #15Whisper Large v3via Together AI192 ms
  16. #16Pulse206 ms
  17. #17Grok STT222 ms
  18. #18resonant-1250 ms
  19. #19Default262 ms
  20. #20Linden 1275 ms
  21. #21Gemini 3.5 Transcribe Live323 ms
Show all 29 models
  1. #22Daytona STT399 ms
  2. #23Voxtral Mini Transcribe Realtime 2602431 ms
  3. #24Solaria 1534 ms
  4. #25GPT Realtime Whisper598 ms
  5. #26GPT-4o mini Transcribe712 ms
  6. #27Chirp 3769 ms
  7. #28GPT-4o Transcribe784 ms
  8. #29Chirp 2910 ms
median of all models · 192 ms
STT models over the last 30 days, ranked on Time to Final Segment.
#ModelHostTTFSWERTTFTSamples
1Qwen3 ASR 1.7bBaseten39 ms946 ms13,669
2STT RT v5Soniox45 ms1519 ms25,416
3Qwen3 ASR FastNari Labs47 ms1752 ms26,672
4STT 1Inworld AI68 ms1166 ms24,769
5Nova 3Deepgram85 ms1436 ms26,023
6Nova 2Deepgram87 ms1418 ms25,978
7Parakeet TDT 0.6B v3Together AI96 ms1222 ms26,478
8Flux MultilingualDeepgram98 ms1166 ms26,026
9FluxDeepgram99 ms1091 ms26,027
10Scribe v2 RealtimeElevenLabs120 ms2154 ms26,660
11Whisper Large v3Baseten132 ms925 ms13,667
12Universal 3.6 ProAssemblyAI134 ms1074 ms26,636
13Ink 2Cartesia149 ms1985 ms26,673
14Universal 3.5 ProAssemblyAI165 ms1023 ms26,674
15Whisper Large v3Together AI192 ms1246 ms26,487
16PulseSmallest206 ms1977 ms26,674
17Grok STTSpaceXAI222 ms—26,596
18resonant-1Reson8250 ms—26,673
19DefaultGradium262 ms1991 ms26,671
20Linden 1Speechmatics275 ms1539 ms26,615
21Gemini 3.5 Transcribe LiveGemini323 ms1574 ms26,580
22Daytona STTGuava399 ms5690 ms6,088
23Voxtral Mini Transcribe Realtime 2602Mistral431 ms1858 ms25,852
24Solaria 1Gladia534 ms1678 ms26,491
25GPT Realtime WhisperOpenAI598 ms1825 ms26,641
26GPT-4o mini TranscribeOpenAI712 ms—26,641
27Chirp 3Google769 ms5958 ms26,659
28GPT-4o TranscribeOpenAI784 ms—26,643
29Chirp 2Google910 ms6101 ms26,670
—Nemotron 3.5 ASR StreamingTogether AI—1547 ms26,536
—Universal StreamingAssemblyAI—1510 ms26,644
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: 6th of 31 on Word Error Rate.

Latency vs accuracy

TTFS vs WEREach point is one measured model · Gemini 3.5 Transcribe Live highlighted · 30-day averagesTime to Final Segment against Word Error Rate for every measured STT model, with Gemini 3.5 Transcribe Live highlighted.
Qwen3 ASR 1.7b — 39 ms, 2.9%STT RT v5 — 45 ms, 4.8%Qwen3 ASR Fast — 47 ms, 2.6%STT 1 — 68 ms, 6.8%Nova 3 — 85 ms, 5.3%Nova 2 — 87 ms, 6.7%Parakeet TDT 0.6B v3 — 96 ms, 11.1%Flux Multilingual — 98 ms, 7.5%Flux — 99 ms, 5.8%Scribe v2 Realtime — 120 ms, 4.3%Whisper Large v3 via Baseten — 132 ms, 4.4%Universal 3.6 Pro — 134 ms, 2.2%Ink 2 — 149 ms, 3.6%Universal 3.5 Pro — 165 ms, 2.4%Whisper Large v3 via Together AI — 192 ms, 7.4%Pulse — 206 ms, 4.6%Grok STT — 222 ms, 4.2%resonant-1 — 250 ms, 2.6%Default — 262 ms, 8.7%Linden 1 — 275 ms, 3.9%Daytona STT — 399 ms, 4.9%Voxtral Mini Transcribe Realtime 2602 — 431 ms, 5.4%Solaria 1 — 534 ms, 6.1%GPT Realtime Whisper — 598 ms, 4.1%GPT-4o mini Transcribe — 712 ms, 3.9%Chirp 3 — 769 ms, 3.2%GPT-4o Transcribe — 784 ms, 3.7%Chirp 2 — 910 ms, 3.9%Gemini 3.5 Transcribe Live — 323 ms, 3.0%

Where the errors come from

WER compositionGemini 3.5 Transcribe Live's Word Error Rate split by error type · 30-day averageGemini 3.5 Transcribe Live's WER split into substitutions, deletions and insertions.
  • Gemini 3.5 Transcribe Live3.0%
SubstitutionsDeletionsInsertions

Averages and tail latency

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

Gemini 3.5 Transcribe Live latency distributionMilliseconds · shared axis across metrics · last 30 daysGemini 3.5 Transcribe Live's latency percentiles per metric: p25–p75 band, p50 tick, whisker to max.
  • Time to Final Segmentp50 301 ms · Max 4965 ms
  • Time to First Tokenp50 1523 ms · Max 12993 ms
band p25–p75 · tick p50 · whisker to max with p90 and p95 stops
Average and percentile values per metric, with the number of samples behind each row.
MetricAveragep25p50p75p90p95MaxSamples
Time to Final Segment323 ms264 ms301 ms369 ms445 ms479 ms4965 ms26,580
Word Error Rate3.0%0.0%0.0%4.0%11.1%17.4%100.0%26,592
Time to First Token1574 ms1285 ms1523 ms1834 ms2152 ms2461 ms12993 ms26,592

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 daysGemini 3.5 Transcribe Live's daily median Time to Final Segment over the last 30 days.
Gemini 3.5 Transcribe Live · Sep 17: 273 msGemini 3.5 Transcribe Live · Sep 18: 259 msGemini 3.5 Transcribe Live · Sep 19: 276 msGemini 3.5 Transcribe Live · Sep 20: 263 msGemini 3.5 Transcribe Live · Sep 21: 264 msGemini 3.5 Transcribe Live · Sep 22: 260 msGemini 3.5 Transcribe Live · Sep 23: 256 msGemini 3.5 Transcribe Live · Sep 24: 272 msGemini 3.5 Transcribe Live · Sep 25: 325 msGemini 3.5 Transcribe Live · Sep 26: 331 msGemini 3.5 Transcribe Live · Sep 27: 337 msGemini 3.5 Transcribe Live · Sep 28: 311 msGemini 3.5 Transcribe Live · Sep 29: 336 msGemini 3.5 Transcribe Live · Sep 30: 382 msGemini 3.5 Transcribe Live · Oct 1: 361 msGemini 3.5 Transcribe Live · Oct 2: 303 msGemini 3.5 Transcribe Live · Oct 3: 302 msGemini 3.5 Transcribe Live · Oct 4: 313 msGemini 3.5 Transcribe Live · Oct 5: 279 msGemini 3.5 Transcribe Live · Oct 6: 273 msGemini 3.5 Transcribe Live · Oct 7: 289 msGemini 3.5 Transcribe Live · Oct 8: 349 msGemini 3.5 Transcribe Live · Oct 9: 323 msGemini 3.5 Transcribe Live · Oct 10: 323 msGemini 3.5 Transcribe Live · Oct 11: 350 msGemini 3.5 Transcribe Live · Oct 12: 322 msGemini 3.5 Transcribe Live · Oct 13: 317 msGemini 3.5 Transcribe Live · Oct 14: 378 msGemini 3.5 Transcribe Live · Oct 15: 375 msGemini 3.5 Transcribe Live · Oct 16: 337 ms
Word Error Rate — daily averageDaily average · UTC daysGemini 3.5 Transcribe Live's daily Word Error Rate over the last 30 days.
Gemini 3.5 Transcribe Live · Sep 17: 4.5%Gemini 3.5 Transcribe Live · Sep 18: 5.2%Gemini 3.5 Transcribe Live · Sep 19: 3.8%Gemini 3.5 Transcribe Live · Sep 20: 3.5%Gemini 3.5 Transcribe Live · Sep 21: 4.1%Gemini 3.5 Transcribe Live · Sep 22: 4.6%Gemini 3.5 Transcribe Live · Sep 23: 3.3%Gemini 3.5 Transcribe Live · Sep 24: 5.3%Gemini 3.5 Transcribe Live · Sep 25: 4.6%Gemini 3.5 Transcribe Live · Sep 26: 5.1%Gemini 3.5 Transcribe Live · Sep 27: 3.4%Gemini 3.5 Transcribe Live · Sep 28: 6.9%Gemini 3.5 Transcribe Live · Sep 29: 3.7%Gemini 3.5 Transcribe Live · Sep 30: 4.9%Gemini 3.5 Transcribe Live · Oct 1: 5.0%Gemini 3.5 Transcribe Live · Oct 2: 3.8%Gemini 3.5 Transcribe Live · Oct 3: 5.5%Gemini 3.5 Transcribe Live · Oct 4: 4.7%Gemini 3.5 Transcribe Live · Oct 5: 4.9%Gemini 3.5 Transcribe Live · Oct 6: 5.5%Gemini 3.5 Transcribe Live · Oct 7: 3.6%Gemini 3.5 Transcribe Live · Oct 8: 4.2%Gemini 3.5 Transcribe Live · Oct 9: 4.6%Gemini 3.5 Transcribe Live · Oct 10: 3.5%Gemini 3.5 Transcribe Live · Oct 11: 3.5%Gemini 3.5 Transcribe Live · Oct 12: 5.0%Gemini 3.5 Transcribe Live · Oct 13: 4.2%Gemini 3.5 Transcribe Live · Oct 14: 4.0%Gemini 3.5 Transcribe Live · Oct 15: 3.9%Gemini 3.5 Transcribe Live · Oct 16: 4.3%

Time to Final Segment by dataset

Gemini 3.5 Transcribe Live by test conditionMilliseconds · lower is better · best condition firstGemini 3.5 Transcribe Live's Time to Final Segment on each benchmark dataset.
  1. WildASR accents297 ms
  2. WildASR clipping312 ms
  3. WildASR phone codec321 ms
  4. WildASR clean322 ms
  5. WildASR far-field322 ms
  6. WildASR noise gaps325 ms
  7. PipeCat (production)328 ms
Time to Final Segment per dataset, with the number of samples behind each figure.
DatasetTTFSSamples
ProductionPipeCat328 ms14,000
AccentsWildASR297 ms1,402
CleanWildASR322 ms5,610
ClippingWildASR312 ms1,364
Far-fieldWildASR322 ms1,400
Noise gapsWildASR325 ms1,401
Phone codecWildASR321 ms1,403

Strongest condition: WildASR accents at 297 ms · weakest: PipeCat (production) at 328 ms.

How fast is Gemini 3.5 Transcribe Live?

On Gemini, Gemini 3.5 Transcribe Live measures mean 323 ms time to final segment (21st of 29) and mean 1574 ms time to first token (16th of 27). Last measured 2026-10-16.

How accurate is Gemini 3.5 Transcribe Live?

On Gemini, Gemini 3.5 Transcribe Live measures 3.0% word error rate (6th of 31). Last measured 2026-10-16.

Who hosts Gemini 3.5 Transcribe Live?

Gemini 3.5 Transcribe Live is created by Google and served by Gemini. Coval measures each hosted endpoint separately.

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