Skip to content

Data read October 27, 2026

Proprietary TTS APIs in 2026: 75 models

75 models have Proprietary access. 21 of the 75 are in the multilingual speed ranking; the others are not tagged multilingual by Coval or are not measured by it.

The access type comes from Coval and Artificial Analysis.

How the rows are ordered. Ranked models first, in ranking order (speed tier, then P95, ties by median); the 54 models without a rank follow alphabetically.

Proprietary TTS APIs in 2026: 75 models
Rank Model Provider Speed tier From the Coval P95 P95 (ms) 1 turn in 20 is slower Listener Elo Artificial Analysis, with 95% interval Price per 1M characters List price; a range when sources differ
1 = Gradium TTS GradiumHeadroom104.31,174 ±13$36 to $57.80
1 = Inworld Realtime TTS-2 Flash InworldHeadroom100.91,218 ±13$10.40 to $15
5 SpeechifyAI Simba 3.0 SpeechifyFast enough149.41,116 ±11$6.60 to $10
6 Inworld Realtime TTS-2 InworldFast enough194.71,267 ±14$15 to $25
7 Palabra TTS v1 PalabraFast enough214.3n/a$12
8 ElevenLabs Eleven v4 Turbo ElevenLabsFast enough238.61,333 ±16$40
9 ElevenLabs Flash v2.5 ElevenLabsFast enough255.11,076 ±11$40 to $50
10 = Soniox TTS RT v1 SonioxVisible delay319.0n/a$13 to $14.10
10 = Soniox TTS Real-Time v2 SonioxVisible delay324.71,180 ±11$13 to $14.81
12 = Murf AI Falcon 2 Murf AIVisible delay363.51,154 ±12$10
12 = Rime Mist v3 RimeVisible delay359.9n/a$30
14 Cartesia Sonic 3.6 CartesiaVisible delay374.41,292 ±14$39.20 to $50
15 Cartesia Sonic 3.5 CartesiaVisible delay390.51,190 ±11$39.20 to $50
16 = Deepdub PhantomZ 4.0 DeepdubVisible delay430.0n/an/a
16 = Rime Coda RimeVisible delay428.11,054 ±12$50
18 SpaceXAI Grok TTS xAIVisible delay473.91,155 ±13$15
19 Fish Audio S2.1 Pro Fish AudioVisible delay533.01,184 ±13$15
20 Smallest.ai Lightning V3.1 Pro (Jul 2026) Smallest.aiVisible delay703.31,169 ±12$19.50
21 Alibaba Qwen3 TTS Flash Realtime Alibaba (Qwen)Visible delay785.8n/a$13
22 Google Chirp 3: HD GoogleVisible delay917.21,061 ±12$30
23 Fish Audio S2.1 Pro Free Fish AudioVisible delay1,137.9n/an/a
not ranked Alibaba Qwen-Audio-3.0-TTS-Plus Alibaba (Qwen)n/an/a1,269 ±13$19.30
not ranked Alibaba Qwen-Audio-3.1-TTS-Plus Alibaba (Qwen)n/an/a1,306 ±15$19.30
not ranked Alibaba Qwen3 TTS Alibaba (Qwen)n/an/a929 ±13$13.80
not ranked Alibaba Qwen3 TTS Flash Alibaba (Qwen)n/an/a939 ±14$10 to $11
not ranked Amazon Polly Generative Amazonn/an/a1,063 ±11$30
not ranked Amazon Polly Long-Form Amazonn/an/a1,029 ±13$100
not ranked Amazon Polly Neural Amazonn/an/a894 ±14$16
not ranked Amazon Polly Standard Amazonn/an/a829 ±16$4
not ranked async Flash v1.0 Asyncn/an/a1,049 ±11$10.10
not ranked async Flash v1.5 Asyncn/an/a1,137 ±12$10.10
not ranked async Pro v1.0 Asyncn/an/a1,127 ±12$20.20
not ranked Bland Speech v3 Bland AIn/an/a1,033 ±12$40
not ranked Cartesia Sonic 3 Cartesian/an/a1,054 ±11$49 to $50
not ranked Deepgram Aura 2 DeepgramVisible delay525.4n/a$30
not ranked Deepgram Aura 2 En DeepgramVisible delay985.7n/an/a
not ranked Deepgram Flux TTS DeepgramFast enough267.01,052 ±13$45
not ranked ElevenLabs Eleven v3 ElevenLabsn/an/a1,184 ±10$100
not ranked ElevenLabs Eleven v3 Conversational ElevenLabsn/an/a1,213 ±12$50
not ranked ElevenLabs Eleven v4 ElevenLabsn/an/a1,333 ±16$80
not ranked ElevenLabs Multilingual v2 ElevenLabsn/an/a1,093 ±10$100
not ranked ElevenLabs Turbo v2.5 ElevenLabsn/an/a1,093 ±10$50
not ranked Fish Audio OpenAudio S1 Fish Audion/an/a1,084 ±11$15
not ranked Fluxions vui FluxionsHeadroom75.0n/a$9
not ranked Google Gemini 2.5 Flash Lite TTS Googlen/an/a1,087 ±11$9.20
not ranked Google Gemini 2.5 Flash TTS (Dec 2025) Googlen/an/a1,067 ±12$9.20
not ranked Google Gemini 2.5 Pro (Dec 2025) Googlen/an/a1,040 ±12$18.30
not ranked Google Gemini 3.1 Flash TTS Googlen/an/a1,212 ±11$18.30 to $33.30
not ranked Google Gemini 3.8 Flash TTS Googlen/an/a1,280 ±14$15 to $16.50
not ranked Google Gemini 3.8 Flash-Lite TTS Googlen/an/a1,256 ±13$10 to $11
not ranked Gradium TTS (Aug 2026) Gradiumn/an/a1,151 ±12$47.20
not ranked Hume AI Octave 2 Hume AIn/an/a1,056 ±12$87.50 to $100
not ranked Hume AI Octave TTS Hume AIn/an/a1,036 ±11$87.50 to $100
not ranked KugelAudio Kugel 3 KugelAudion/an/a924 ±13n/a
not ranked Maya 2 Flash Maya Researchn/an/a1,065 ±13n/a
not ranked Maya 2 Global Maya Researchn/an/a1,047 ±13n/a
not ranked Microsoft Azure HD 2.5 Microsoftn/an/a1,142 ±11$22
not ranked Microsoft Azure Neural Microsoftn/an/a1,032 ±20$15
not ranked MiniMax Speech 2.6 HD MiniMaxn/an/a1,126 ±11$100
not ranked MiniMax Speech 2.6 Turbo MiniMaxn/an/a1,120 ±11$60
not ranked MiniMax Speech 2.8 HD MiniMaxn/an/a1,174 ±10$100
not ranked MiniMax Speech 2.8 Turbo MiniMaxn/an/a1,153 ±10$60
not ranked MiniMax Speech-02-HD MiniMaxn/an/a1,108 ±11$100
not ranked MiniMax Speech-02-Turbo MiniMaxn/an/a1,089 ±11$60
not ranked Mistral Voxtral 3 TTS Mistraln/an/a1,221 ±14$13
not ranked Neuphonic TTS Neuphonicn/an/a940 ±14$20.80
not ranked OpenAI GPT-Realtime-2 OpenAIn/an/a1,073 ±12$191.60
not ranked OpenAI TTS-1 OpenAIn/an/a1,088 ±10$15
not ranked OpenAI TTS-1 HD OpenAIn/an/a1,106 ±11$30
not ranked Out of Set Airy TTS Out of SetVisible delay453.9n/a$3.67
not ranked Rime Mist V2 Rimen/an/a902 ±15$31.10
not ranked Smallest.ai Lightning v3.1 Smallest.ain/an/a1,020 ±12$17.50 to $25
not ranked SpeechifyAI Simba 3.2 SpeechifyFast enough161.11,249 ±12$6.60 to $10
not ranked StepFun Step TTS 2 (Mar 2026) StepFunn/an/a1,135 ±11$40
not ranked StepFun StepAudio 2.5 TTS (Aug 2026) StepFunn/an/a1,212 ±13$85

Frequently asked questions

Which TTS APIs have Proprietary access?
Starting with the best-ranked, Gradium TTS, Inworld Realtime TTS-2 Flash, SpeechifyAI Simba 3.0, and Inworld Realtime TTS-2; 75 in total.
How many are ranked?
21 of the 75 are in the multilingual speed ranking; the others are not tagged multilingual by Coval or are not measured by it.