Voice AI BenchmarksBy

PROVIDERLAST 30 DAYS

official resource

Smallest voice AI models and benchmarks

Smallest (Smallest, smallest.ai) lists 2 STT and TTS models in Coval. Fastest dated mean latency over 30 days: STT: Pulse at 205 ms TTFS, with 4.6% WER. TTS: Lightning v3.1 Pro at 392 ms TTFA, with 1.5% WER. Last measured .

Smallest.ai develops real-time speech recognition and synthesis.

Measured models
2
STTTTS
Best STT model#16 / 29
205ms
Pulse
Best TTS model#26 / 32
392ms
Lightning v3.1 Pro

Overview

Smallest.ai's Pulse recognizer and Lightning synthesizer are both built and served in-house.

Every model below is measured daily on the same fixed inputs and ranked against the full field, never blended into a company score.

Model lineup

Speech-to-Text
Pulse
Text-to-Speech
Lightning v3.1 Pro

Speech-to-Text

Full STT dashboard

Ranked on Time to Final Segment against 29 measured models.

Time to Final Segmentms · lower is better · Smallest models markedEvery measured STT model on Time to Final Segment, with Smallest's models highlighted.
  1. #1Qwen3 ASR 1.7b39 ms
  2. #2STT RT v544 ms
  3. #3Qwen3 ASR Fast47 ms
  4. #4STT 167 ms
  5. #5Nova 385 ms
  6. #6Nova 287 ms
  7. #7Parakeet TDT 0.6B v397 ms
  8. #16Pulse205 ms
leaders plus Smallest models · 21 other models in the full table
Word Error Rate% · lower is better · Smallest models markedEvery measured STT model on Word Error Rate, with Smallest's models highlighted.
  1. #1Universal 3.6 Pro2.1%
  2. #2Universal 3.5 Pro2.4%
  3. #3Qwen3 ASR Fast2.6%
  4. #4resonant-12.6%
  5. #5Qwen3 ASR 1.7b2.9%
  6. #6Gemini 3.5 Transcribe Live3.0%
  7. #7Chirp 33.2%
  8. #17Pulse4.6%
leaders plus Smallest models · 23 other models in the full table
Time to First Tokenms · lower is better · Smallest models markedEvery measured STT model on Time to First Token, with Smallest's models highlighted.
  1. #1Whisper Large v3via Baseten924 ms
  2. #2Qwen3 ASR 1.7b946 ms
  3. #3Universal 3.5 Pro1025 ms
  4. #4Universal 3.6 Pro1071 ms
  5. #5Flux1091 ms
  6. #6Flux Multilingual1165 ms
  7. #7STT 11199 ms
  8. #22Pulse1976 ms
leaders plus Smallest models · 19 other models in the full table
TTFS vs WER — Smallest vs every measured modelEach point is one measured STT model · Smallest models highlighted · 30-day averagesTime to Final Segment against Word Error Rate for every measured STT model, with Smallest's models highlighted.
Qwen3 ASR 1.7b — 39 ms, 2.9%STT RT v5 — 44 ms, 4.8%Qwen3 ASR Fast — 47 ms, 2.6%STT 1 — 67 ms, 7.1%Nova 3 — 85 ms, 5.3%Nova 2 — 87 ms, 6.7%Parakeet TDT 0.6B v3 — 97 ms, 11.1%Flux Multilingual — 98 ms, 7.5%Flux — 98 ms, 5.8%Scribe v2 Realtime — 121 ms, 4.4%Whisper Large v3 via Baseten — 131 ms, 4.5%Universal 3.6 Pro — 135 ms, 2.1%Ink 2 — 149 ms, 3.6%Universal 3.5 Pro — 170 ms, 2.4%Whisper Large v3 via Together AI — 196 ms, 7.4%Grok STT — 222 ms, 4.1%resonant-1 — 252 ms, 2.6%Default — 262 ms, 8.8%Linden 1 — 275 ms, 3.9%Gemini 3.5 Transcribe Live — 319 ms, 3.0%Daytona STT — 399 ms, 4.9%Voxtral Mini Transcribe Realtime 2602 — 416 ms, 5.4%GPT Realtime Whisper — 599 ms, 4.0%Solaria 1 — 614 ms, 6.1%GPT-4o mini Transcribe — 711 ms, 3.9%Chirp 3 — 768 ms, 3.2%GPT-4o Transcribe — 784 ms, 3.7%Chirp 2 — 910 ms, 3.9%Pulse — 205 ms, 4.6%
Benchmarked models with their Time to Final Segment over the last 30 days.
ModelHostTTFSRank
PulseSmallest205 ms16th of 29

Text-to-Speech

Full TTS dashboard

Ranked on Time to First Audio against 32 measured models.

Time to First Audioms · lower is better · Smallest models markedEvery measured TTS model on Time to First Audio, with Smallest's models highlighted.
  1. #1vui59 ms
  2. #2Default63 ms
  3. #3Qwen3 TTS Fast68 ms
  4. #4TTS Flash 291 ms
  5. #5Qwen3 TTS 1.7b107 ms
  6. #6Simba 3.0111 ms
  7. #7Palabra TTS121 ms
  8. #26Lightning v3.1 Pro392 ms
leaders plus Smallest models · 24 other models in the full table
Word Error Rate% · lower is better · Smallest models markedEvery measured TTS model on Word Error Rate, with Smallest's models highlighted.
  1. #1Sonic 3.61.5%
  2. #2S2.1 Pro Free1.5%
  3. #3S2.1 Pro1.5%
  4. #4Simba 3.01.5%
  5. #5Lightning v3.1 Pro1.5%
  6. #6Sonic 3.51.5%
  7. #7Chirp 3 HD1.5%
leaders plus Smallest models · 25 other models in the full table
TTFA vs WER — Smallest vs every measured modelEach point is one measured TTS model · Smallest models highlighted · 30-day averagesTime to First Audio against Word Error Rate for every measured TTS model, with Smallest's models highlighted.
vui — 59 ms, 2.1%Default — 63 ms, 1.9%Qwen3 TTS Fast — 68 ms, 1.8%TTS Flash 2 — 91 ms, 1.7%Qwen3 TTS 1.7b — 107 ms, 1.8%Simba 3.0 — 111 ms, 1.5%Palabra TTS — 121 ms, 2.6%Simba 3.2 — 126 ms, 1.9%drama3-preview — 168 ms, 1.9%TTS 2 — 190 ms, 1.6%Flash v2.5 — 192 ms, 1.8%Eleven v4 Turbo — 200 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%Phantom Z 3.4 conversational — 287 ms, 1.8%Sonic 3.5 — 293 ms, 1.5%Falcon 2 — 296 ms, 1.7%Aura 2 — 297 ms, 1.8%Sonic 3.6 — 299 ms, 1.5%Coda — 332 ms, 9.5%S2.1 Pro — 342 ms, 1.5%Grok TTS — 362 ms, 1.6%S1 — 390 ms, 1.8%Daytona TTS — 486 ms, 1.6%Aura 2 En — 568 ms, 2.1%Chirp 3 HD — 599 ms, 1.5%Qwen3 TTS Flash Realtime — 718 ms, 3.0%S2.1 Pro Free — 848 ms, 1.5%GPT-4o mini TTS — 2582 ms, 1.8%Lightning v3.1 Pro — 392 ms, 1.5%
Benchmarked models with their Time to First Audio over the last 30 days.
ModelHostTTFARank
Lightning v3.1 ProSmallest392 ms26th of 32

How fast are Smallest's STT and TTS models?

Pulse measures mean 205 ms time to final segment (16th of 29) among STT systems. Last measured 2026-10-15. Lightning v3.1 Pro measures mean 392 ms time to first audio (26th of 32) among TTS systems. Last measured 2026-10-15.

How accurate are Smallest's STT and TTS models?

Pulse measures 4.6% word error rate (17th of 31) among STT systems. Last measured 2026-10-15. Lightning v3.1 Pro measures 1.5% word error rate (5th of 32) among TTS systems. Last measured 2026-10-15.

Which Smallest model is fastest?

Its fastest dated STT result is Pulse at mean 205 ms time to final segment (16th of 29) among STT systems, with 4.6% WER. Last measured 2026-10-15. Its fastest dated TTS result is Lightning v3.1 Pro at mean 392 ms time to first audio (26th of 32) among TTS systems, with 1.5% WER. Last measured 2026-10-15.

Limits of this comparison

Coval measures Pulse on incoming audio and Lightning v3.1 Pro on fixed text prompts.

  • Coval does not publish one combined Smallest.ai pipeline score or evaluate every voice-cloning capability.

Official resources

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