Independent Text-to-Speech (TTS) benchmark
An independent benchmark of text-to-speech providers on Time to First Audio (TTFA) and Word Error Rate (WER), measured under production-realistic conditions by Coval. Fastest is vui (101 ms median TTFA), lowest word error rate is tts-rt-v1 (3.7%). Mirrored here with attribution; last 7d.
Which text-to-speech API is best?
There is no single “best”. The measured split: vui (Fluxions) is the fastest at 101 ms median TTFA, while tts-rt-v1 (Soniox) is the most accurate at 3.7% WER. Route by the axis your workload is constrained by:
| Your constraint | Pick by | Current leader |
|---|---|---|
| Real-time voice agent — the reply must start speaking instantly | lowest median TTFA | vui Fluxions · 101 ms |
| Content fidelity — names, numbers, and scripts must come out exactly right | lowest WER | tts-rt-v1 Soniox · 3.7% |
Not measured here: voice quality / naturalness and price: treat those as vendor claims until measured. Full ranking below.
Run by Coval, a voice-AI evaluation platform
This benchmark is produced by Coval, which sells voice-agent evaluation infrastructure. Every number here is Coval's, measured on a pinned dataset under production-realistic conditions; the methodology, runner code, and inputs are open-source (Apache-2.0) and reproducible by re-running the suite. Openbenchmarks mirrors these results with attribution.
These figures are a rolling 7d window, not one job. A Coval run ID is not printed here because it does not identify this table. The table hash is SHA-256 of the numbers on this page. The dataset hash is the pinned audio set.
Text-to-Speech models ranked by Time to First Audio
Every model runs the same pinned dataset under the same conditions. TTFA (median) is the headline latency; WER is accuracy (lower is better on both). Ranked fastest-first, last 7d:
| # | Model | TTFA median | TTFA p95 | WER | Samples |
|---|---|---|---|---|---|
| 1 | vui † Fluxions | 101 ms | 162 ms | 6.2% | 2,304 |
| 2 | palabra-tts-v1 Palabra | 105 ms | 154 ms | 6.0% | 3,226 |
| 3 | inworld-tts-2-flash Inworld AI | 113 ms | 174 ms | 4.8% | 3,350 |
| 4 | inworld-tts-2 Inworld AI | 162 ms | 243 ms | 4.6% | 3,351 |
| 5 | blizzard LMNT | 217 ms | 405 ms | 7.7% | 3,350 |
| 6 | tts-rt-v2 Soniox | 228 ms | 275 ms | 3.9% | 3,349 |
| 7 | tts-rt-v1 Soniox | 229 ms | 283 ms | 3.7% | 3,350 |
| 8 | mistv3 Rime | 252 ms | 284 ms | 5.9% | 3,351 |
| 9 | sonic-3.5 Cartesia | 272 ms | 353 ms | 5.9% | 3,351 |
| 10 | aura-2-thalia-en Deepgram | 286 ms | 539 ms | 5.4% | 3,344 |
| 11 | coda Rime | 312 ms | 404 ms | 5.1% | 3,351 |
| 12 | s2.1-pro Fish Audio | 316 ms | 513 ms | 4.7% | 3,351 |
| 13 | lightning_v3.1_pro Smallest | 331 ms | 614 ms | 4.4% | 3,339 |
| 14 | s2.1-pro-free Fish Audio | 350 ms | 1,331 ms | 4.6% | 3,317 |
| 15 | grok-tts xAI | 383 ms | 530 ms | 4.9% | 3,347 |
| 16 | default Gradium | 386 ms | 485 ms | 4.6% | 3,350 |
| 17 | s1 Fish Audio | 387 ms | 545 ms | 4.9% | 3,351 |
| 18 | simba-3.2 Speechify | 404 ms | 654 ms | 4.3% | 3,348 |
| 19 | simba-3.0 Speechify | 414 ms | 585 ms | 5.2% | 3,351 |
| 20 | speech-2.8-turbo † Minimax | 426 ms | 507 ms | 4.2% | 258 |
| 21 | speech-2.8-hd † Minimax | 471 ms | 588 ms | 3.9% | 258 |
| 22 | chirp-3-hd | 493 ms | 887 ms | 5.1% | 3,350 |
| 23 | falcon-2 Murf | 557 ms | 773 ms | 5.4% | 3,351 |
| 24 | qwen3-tts-flash-realtime Alibaba | 697 ms | 816 ms | 8.8% | 3,351 |
| 25 | gpt-4o-mini-tts OpenAI | 775 ms | 3,907 ms | 4.8% | 3,350 |
Numbers are point-in-time against Coval's pinned dataset and refresh continuously. Full charts, distributions, and windows (24h / 7d / 30d) at benchmarks.coval.ai →
Compare two text-to-speech providers directly
Provider vs provider on the same measured data: best model per axis, winner per axis:
ElevenLabs vs Cartesia → · ElevenLabs vs OpenAI → · Cartesia vs Deepgram →
Text-to-Speech benchmark: common questions
Which text-to-speech API is the fastest?
vui (Fluxions) has the lowest median Time to First Audio in the current 7d window: 101 ms. Next is palabra-tts-v1 (Palabra) at 105 ms. Latency is TTFA: time from sending the synthesis request to the first byte of audio coming back — the pause a caller hears before the voice starts speaking. Lower is better.
Which text-to-speech API is the most accurate?
tts-rt-v1 (Soniox) has the lowest average Word Error Rate: 3.7%. Share of words wrong in the synthesized audio: the audio is transcribed back with a reference speech-to-text model and compared to the input text, which catches skipped, garbled, and hallucinated words. Lower is better.
What is Time to First Audio (TTFA)?
Time from sending the synthesis request to the first byte of audio coming back — the pause a caller hears before the voice starts speaking. Lower is better.
What is Word Error Rate (WER)?
Share of words wrong in the synthesized audio: the audio is transcribed back with a reference speech-to-text model and compared to the input text, which catches skipped, garbled, and hallucinated words. Lower is better.
Does this benchmark measure voice quality or price?
No. It measures latency (TTFA) and Word Error Rate on identical inputs. Voice quality / naturalness (how human the voice sounds) and price are not measured; treat those as vendor claims until measured.
Is this TTS benchmark independent? Who runs it?
Coval sells voice-agent evaluation infrastructure. The methodology, runner code, and pinned dataset are open-source (Apache-2.0) and re-runnable end to end; Openbenchmarks mirrors the results with attribution.
How fresh is this data?
The table is a rolling 7d aggregate, not a single Coval job. Coval re-runs continuously (roughly every 30 minutes); this page mirrors the window and shows a hash of the figures on this page. Last pull 2026-08-25 06:31 UTC. A Coval run ID is not printed next to the table because it does not identify these numbers.