Voice AI BenchmarksBy

INWORLD AITEXT-TO-SPEECH6,758 SAMPLES / 30 DAYSLAST RUN OCT 14, 2026, 08:00 UTC

official resource

TTS 2 text-to-speech benchmarks

TTS 2, hosted by Inworld AI, measures mean 191 ms time to first audio (11th of 32) and 1.6% word error rate (10th of 32) among TTS systems. Results cover the last 30 days. Coval benchmarks hosted endpoints daily using a consistent evaluation methodology. Last measured .

Inworld TTS 2 is Inworld's full synthesis model for interactive applications.

Time to First Audio#11 / 32
191ms
TTFA Network Roundtrip#12 / 32
137ms
TTFA Leading Silence#14 / 32
54ms
Word Error Rate#10 / 32
1.6%

Overview

Inworld offers multilingual voices, cloning and emotion controls. The standard and Flash endpoints are measured separately.

TTS 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
Inworld AI
Hosted by
Inworld AI
Source
Official API
Licensing
Proprietary
Deployment
Cloud
Region
US
Features
Emotion control, Multilingual, Voice cloning

How TTS 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 TTS 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-preview165 ms
  10. #10Flash v2.5191 ms
  11. #11TTS 2191 ms
  12. #12Eleven v4 Turbo201 ms
Show all 32 models
  1. #13Flux TTS205 ms
  2. #14TTS RT v1254 ms
  3. #15TTS Rt v2256 ms
  4. #16Mist v3287 ms
  5. #17Sonic 3.5293 ms
  6. #18Phantom Z 3.4 conversational295 ms
  7. #19Aura 2297 ms
  8. #20Sonic 3.6299 ms
  9. #21Falcon 2302 ms
  10. #22Coda332 ms
  11. #23S2.1 Pro341 ms
  12. #24Grok TTS362 ms
  13. #25S1389 ms
  14. #26Lightning v3.1 Pro392 ms
  15. #27Daytona TTS486 ms
  16. #28Aura 2 En572 ms
  17. #29Chirp 3 HD598 ms
  18. #30Qwen3 TTS Flash Realtime718 ms
  19. #31S2.1 Pro Free850 ms
  20. #32GPT-4o mini TTS2582 ms
median of all models · 290 ms
TTS models over the last 30 days, ranked on Time to First Audio.
#ModelHostTTFAWERSamples
1vuiFluxions60 ms1,714
2Gradium TTS Beta 202609Gradium63 ms1,886
3Qwen3 TTS FastNari Labs68 ms1,886
4TTS Flash 2Inworld AI92 ms1,690
5Qwen3 TTS 1.7bBaseten107 ms1,248
6Simba 3.0SpeechifyAI111 ms1,885
7Palabra TTS v1Palabra119 ms1,549
8Simba 3.2SpeechifyAI126 ms1,882
9Drama 3 Preview Us EastFish Audio165 ms132
10Flash v2.5ElevenLabs191 ms1,877
11TTS 2Inworld AI191 ms1,690
12Eleven v4 TurboElevenLabs201 ms1,196
13Flux TTSDeepgram205 ms1,749
14TTS RT v1Soniox254 ms1,590
15TTS Rt v2Soniox256 ms1,587
16Mist v3Rime287 ms602
17Sonic 3.5Cartesia293 ms1,886
18Phantom Z 3.4 conversationalDeepdub295 ms1,621
19Aura 2Deepgram297 ms1,818
20Sonic 3.6Cartesia299 ms1,886
21Falcon 2Murf302 ms1,885
22CodaRime332 ms602
23S2.1 ProFish Audio341 ms1,886
24Grok TTSSpaceXAI362 ms1,881
25S1Fish Audio389 ms1,885
26Lightning v3.1 ProSmallest392 ms1,886
27Daytona TTSGuava486 ms1,152
28Aura 2 EnCloudflare572 ms1,885
29Chirp 3 HDGoogle598 ms1,886
30Qwen3 TTS Flash RealtimeAlibaba718 ms1,885
31S2.1 Pro FreeFish Audio850 ms1,879
32GPT-4o mini TTSOpenAI2582 ms1,822
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: 10th of 32 on Word Error Rate.

Latency vs accuracy

TTFA vs WEREach point is one measured model · TTS 2 highlighted · 30-day averagesTime to First Audio against Word Error Rate for every measured TTS model, with TTS 2 highlighted.
vui — 60 ms, 2.1%Default — 63 ms, 1.9%Qwen3 TTS Fast — 68 ms, 1.8%TTS Flash 2 — 92 ms, 1.6%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 — 165 ms, 2.0%Flash v2.5 — 191 ms, 1.8%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%Aura 2 — 297 ms, 1.8%Sonic 3.6 — 299 ms, 1.5%Falcon 2 — 302 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, 2.0%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 — 2582 ms, 1.8%TTS 2 — 191 ms, 1.6%

Where the errors come from

WER compositionTTS 2's Word Error Rate split by error type · 30-day averageTTS 2's WER split into substitutions, deletions and insertions.
  • TTS 21.6%
SubstitutionsDeletionsInsertions

Averages and tail latency

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

TTS 2 latency distributionMilliseconds · shared axis across metrics · last 30 daysTTS 2's latency percentiles per metric: p25–p75 band, p50 tick, whisker to max.
  • Time to First Audiop50 149 ms · Max 31952 ms
  • TTFA Network Roundtripp50 95 ms · Max 31910 ms
  • TTFA Leading Silencep50 53 ms · Max 137 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 Audio191 ms135 ms149 ms167 ms181 ms192 ms31952 ms1,690
TTFA Network Roundtrip137 ms88 ms95 ms102 ms110 ms121 ms31910 ms1,690
TTFA Leading Silence54 ms42 ms53 ms71 ms80 ms84 ms137 ms1,690
Word Error Rate1.6%0.0%0.0%2.0%7.5%12.5%43.8%1,688

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 daysTTS 2's daily median Time to First Audio over the last 30 days.
TTS 2 · Sep 23: 143 msTTS 2 · Sep 24: 153 msTTS 2 · Sep 25: 149 msTTS 2 · Sep 26: 156 msTTS 2 · Sep 27: 144 msTTS 2 · Sep 28: 151 msTTS 2 · Sep 30: 120 msTTS 2 · Oct 1: 150 msTTS 2 · Oct 2: 156 msTTS 2 · Oct 3: 141 msTTS 2 · Oct 4: 148 msTTS 2 · Oct 5: 151 msTTS 2 · Oct 6: 139 msTTS 2 · Oct 7: 155 msTTS 2 · Oct 8: 161 msTTS 2 · Oct 9: 136 msTTS 2 · Oct 10: 147 msTTS 2 · Oct 11: 150 msTTS 2 · Oct 12: 145 msTTS 2 · Oct 13: 139 msTTS 2 · Oct 14: 155 ms
Word Error Rate — daily averageDaily average · UTC daysTTS 2's daily Word Error Rate over the last 30 days.
TTS 2 · Sep 23: 5.8%TTS 2 · Sep 24: 3.4%TTS 2 · Sep 25: 3.2%TTS 2 · Sep 26: 2.9%TTS 2 · Sep 27: 4.9%TTS 2 · Sep 30: 0.5%TTS 2 · Oct 1: 4.9%TTS 2 · Oct 2: 3.1%TTS 2 · Oct 3: 5.0%TTS 2 · Oct 4: 5.9%TTS 2 · Oct 5: 2.9%TTS 2 · Oct 6: 3.9%TTS 2 · Oct 7: 4.8%TTS 2 · Oct 8: 5.9%TTS 2 · Oct 9: 2.6%TTS 2 · Oct 10: 1.1%TTS 2 · Oct 11: 3.3%TTS 2 · Oct 12: 4.9%TTS 2 · Oct 13: 3.5%TTS 2 · Oct 14: 7.8%

How fast is TTS 2?

On Inworld AI, TTS 2 measures mean 191 ms time to first audio (11th of 32). Last measured 2026-10-14.

How accurate is TTS 2?

On Inworld AI, TTS 2 measures 1.6% word error rate (10th of 32). Last measured 2026-10-14.

Who hosts TTS 2?

TTS 2 is created by Inworld AI and served by Inworld AI. Coval measures each hosted endpoint separately.

Limits of this comparison

TTS 2 is shown beside TTS 2 Flash without assuming the faster-named variant makes the same tradeoffs.

  • The fixed prompt set does not score character consistency, emotional appropriateness or clone similarity.

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