Voice AI BenchmarksBy

DEEPGRAMTEXT-TO-SPEECH7,246 SAMPLES / 30 DAYSLAST RUN OCT 14, 2026, 06:30 UTC

official resource

Aura 2 text-to-speech benchmarks

Aura 2, hosted by Deepgram, measures mean 297 ms time to first audio (19th of 32) and 1.8% word error rate (21st of 32) among TTS systems. Results cover the last 30 days. Coval benchmarks hosted endpoints daily using a consistent evaluation methodology. Last measured .

Aura 2 is Deepgram's real-time text-to-speech model, measured here with its Thalia English voice.

Time to First Audio#19 / 32
297ms
TTFA Network Roundtrip#9 / 32
108ms
TTFA Leading Silence#29 / 32
189ms
Word Error Rate#21 / 32
1.8%

Overview

Aura 2 streams synthesis over Deepgram's real-time API, and voice choice can affect startup silence and intelligibility.

Aura 2 is tested every day on a fixed set of text prompts, measuring how quickly audible speech starts and how intelligible the result is.

Technical specifications

Made by
Deepgram
Hosted by
Deepgram
Source
Official API
Licensing
Proprietary
Deployment
On-prem
Region
US

How Aura 2 ranks

Full TTS dashboard
Time to First Audio — every measured TTS modelMilliseconds · lower is better · 30-day averageEvery TTS model ranked on Time to First Audio, with Aura 2 highlighted.
  1. #1vui60 ms
  2. #2Default63 ms
  3. #3Qwen3 TTS Fast68 ms
  4. #4TTS Flash 292 ms
  5. #5Qwen3 TTS 1.7b107 ms
  6. #6Simba 3.0111 ms
  7. #7Palabra TTS v1119 ms
  8. #8Simba 3.2126 ms
  9. #9drama3-preview161 ms
  10. #10Flash v2.5191 ms
  11. #11TTS 2192 ms
  12. #12Eleven v4 Turbo201 ms
  13. #13Flux TTS205 ms
  14. #14TTS RT v1254 ms
  15. #15TTS Rt v2256 ms
  16. #16Mist v3287 ms
  17. #17Sonic 3.5293 ms
  18. #18Phantom Z 3.4 conversational295 ms
  19. #19Aura 2297 ms
Show all 32 models
  1. #20Sonic 3.6299 ms
  2. #21Falcon 2303 ms
  3. #22Coda332 ms
  4. #23S2.1 Pro341 ms
  5. #24Grok TTS362 ms
  6. #25S1389 ms
  7. #26Lightning v3.1 Pro392 ms
  8. #27Daytona TTS486 ms
  9. #28Aura 2 En572 ms
  10. #29Chirp 3 HD598 ms
  11. #30Qwen3 TTS Flash Realtime718 ms
  12. #31S2.1 Pro Free850 ms
  13. #32GPT-4o mini TTS2583 ms
median of all models · 290 ms
TTS models over the last 30 days, ranked on Time to First Audio.
#ModelHostTTFAWERSamples
1vuiFluxions60 ms1,708
2Gradium TTS Beta 202609Gradium63 ms1,880
3Qwen3 TTS FastNari Labs68 ms1,880
4TTS Flash 2Inworld AI92 ms1,684
5Qwen3 TTS 1.7bBaseten107 ms1,248
6Simba 3.0SpeechifyAI111 ms1,879
7Palabra TTS v1Palabra119 ms1,545
8Simba 3.2SpeechifyAI126 ms1,876
9Drama 3 Preview Us EastFish Audio161 ms126
10Flash v2.5ElevenLabs191 ms1,871
11TTS 2Inworld AI192 ms1,684
12Eleven v4 TurboElevenLabs201 ms1,190
13Flux TTSDeepgram205 ms1,743
14TTS RT v1Soniox254 ms1,585
15TTS Rt v2Soniox256 ms1,583
16Mist v3Rime287 ms602
17Sonic 3.5Cartesia293 ms1,880
18Phantom Z 3.4 conversationalDeepdub295 ms1,615
19Aura 2Deepgram297 ms1,812
20Sonic 3.6Cartesia299 ms1,880
21Falcon 2Murf303 ms1,879
22CodaRime332 ms602
23S2.1 ProFish Audio341 ms1,880
24Grok TTSSpaceXAI362 ms1,875
25S1Fish Audio389 ms1,879
26Lightning v3.1 ProSmallest392 ms1,880
27Daytona TTSGuava486 ms1,152
28Aura 2 EnCloudflare572 ms1,879
29Chirp 3 HDGoogle598 ms1,880
30Qwen3 TTS Flash RealtimeAlibaba718 ms1,879
31S2.1 Pro FreeFish Audio850 ms1,873
32GPT-4o mini TTSOpenAI2583 ms1,816
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

TTFA vs WEREach point is one measured model · Aura 2 highlighted · 30-day averagesTime to First Audio against Word Error Rate for every measured TTS model, with Aura 2 highlighted.
vui — 60 ms, 2.1%Default — 63 ms, 1.9%Qwen3 TTS Fast — 68 ms, 1.8%TTS Flash 2 — 92 ms, 1.7%Qwen3 TTS 1.7b — 107 ms, 1.8%Simba 3.0 — 111 ms, 1.5%Palabra TTS v1 — 119 ms, 2.6%Simba 3.2 — 126 ms, 1.9%drama3-preview — 161 ms, 2.1%Flash v2.5 — 191 ms, 1.8%TTS 2 — 192 ms, 1.6%Eleven v4 Turbo — 201 ms, 1.7%Flux TTS — 205 ms, 1.7%TTS RT v1 — 254 ms, 2.4%TTS Rt v2 — 256 ms, 2.4%Mist v3 — 287 ms, 2.1%Sonic 3.5 — 293 ms, 1.5%Phantom Z 3.4 conversational — 295 ms, 1.8%Sonic 3.6 — 299 ms, 1.5%Falcon 2 — 303 ms, 1.7%Coda — 332 ms, 9.5%S2.1 Pro — 341 ms, 1.5%Grok TTS — 362 ms, 1.6%S1 — 389 ms, 1.8%Lightning v3.1 Pro — 392 ms, 1.5%Daytona TTS — 486 ms, 1.6%Aura 2 En — 572 ms, 2.1%Chirp 3 HD — 598 ms, 1.5%Qwen3 TTS Flash Realtime — 718 ms, 3.0%S2.1 Pro Free — 850 ms, 1.5%GPT-4o mini TTS — 2583 ms, 1.8%Aura 2 — 297 ms, 1.8%

Where the errors come from

WER compositionAura 2's Word Error Rate split by error type · 30-day averageAura 2's WER split into substitutions, deletions and insertions.
  • Aura 21.8%
SubstitutionsDeletionsInsertions

Averages and tail latency

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

Aura 2 latency distributionMilliseconds · shared axis across metrics · last 30 daysAura 2's latency percentiles per metric: p25–p75 band, p50 tick, whisker to max.
  • Time to First Audiop50 259 ms · Max 658 ms
  • TTFA Network Roundtripp50 129 ms · Max 271 ms
  • TTFA Leading Silencep50 126 ms · Max 509 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 First Audio297 ms173 ms259 ms426 ms492 ms520 ms658 ms1,812
TTFA Network Roundtrip108 ms63 ms129 ms139 ms145 ms152 ms271 ms1,812
TTFA Leading Silence189 ms77 ms126 ms320 ms367 ms398 ms509 ms1,812
Word Error Rate1.8%0.0%0.0%2.8%8.0%12.2%64.0%1,810

Last 30 days

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

Time to First Audio — daily p50Line p50 · band p25–p75 · UTC daysAura 2's daily median Time to First Audio over the last 30 days.
Aura 2 · Sep 24: 315 msAura 2 · Sep 25: 305 msAura 2 · Sep 26: 352 msAura 2 · Sep 27: 264 msAura 2 · Sep 28: 422 msAura 2 · Sep 29: 169 msAura 2 · Sep 30: 293 msAura 2 · Oct 1: 369 msAura 2 · Oct 2: 198 msAura 2 · Oct 3: 301 msAura 2 · Oct 4: 342 msAura 2 · Oct 5: 213 msAura 2 · Oct 6: 309 msAura 2 · Oct 7: 299 msAura 2 · Oct 8: 225 msAura 2 · Oct 9: 332 msAura 2 · Oct 10: 282 msAura 2 · Oct 11: 378 msAura 2 · Oct 12: 181 msAura 2 · Oct 13: 311 msAura 2 · Oct 14: 246 ms
Word Error Rate — daily averageDaily average · UTC daysAura 2's daily Word Error Rate over the last 30 days.
Aura 2 · Sep 24: 7.6%Aura 2 · Sep 25: 5.2%Aura 2 · Sep 26: 2.3%Aura 2 · Sep 27: 7.2%Aura 2 · Sep 28: 8.7%Aura 2 · Sep 29: 3.5%Aura 2 · Sep 30: 2.1%Aura 2 · Oct 1: 6.2%Aura 2 · Oct 2: 6.2%Aura 2 · Oct 3: 6.4%Aura 2 · Oct 4: 7.0%Aura 2 · Oct 5: 2.7%Aura 2 · Oct 6: 4.0%Aura 2 · Oct 7: 5.3%Aura 2 · Oct 8: 3.9%Aura 2 · Oct 9: 2.9%Aura 2 · Oct 10: 1.8%Aura 2 · Oct 11: 3.3%Aura 2 · Oct 12: 4.9%Aura 2 · Oct 13: 3.8%Aura 2 · Oct 14: 7.8%

How fast is Aura 2?

On Deepgram, Aura 2 measures mean 297 ms time to first audio (19th of 32). Last measured 2026-10-14.

How accurate is Aura 2?

On Deepgram, Aura 2 measures 1.8% word error rate (21st of 32). Last measured 2026-10-14.

Who hosts Aura 2?

Aura 2 is created by Deepgram and served by Deepgram. Coval measures each hosted endpoint separately.

Limits of this comparison

Results apply specifically to this voice because voice choice can affect leading silence and intelligibility.

  • One voice cannot represent every Aura voice, language or speaking style, so the results stay attached to Thalia English.

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