Skip to content

Data read October 27, 2026

Chinese TTS APIs in 2026: 30 models tested in Chinese

30 models were tested in Chinese by Artificial Analysis or Speko. 9 of the 30 are in the multilingual speed ranking; the others are not tagged multilingual by Coval or are not measured by it. The fastest ranked model tested in Chinese is Inworld Realtime TTS-2 Flash.

No public source publishes naturalness or accuracy for Chinese on its own, and all latency is measured in English from the US East. The page therefore says who was tested in Chinese, and how fast each model is in general.

How the rows are ordered. Ranked models first, in ranking order (speed tier, then P95, ties by median); the 21 models without a rank follow alphabetically.

Chinese TTS APIs in 2026: 30 models tested in Chinese
Rank Model Speed tier From the Coval P95 P95 (ms) 1 turn in 20 is slower Languages tested Listener Elo Artificial Analysis, with 95% interval Price per 1M characters List price; a range when sources differ
1 = Inworld Realtime TTS-2 Flash Headroom100.991,218 ±13$10.40 to $15
6 Inworld Realtime TTS-2 Fast enough194.7121,267 ±14$15 to $25
8 ElevenLabs Eleven v4 Turbo Fast enough238.691,333 ±16$40
9 ElevenLabs Flash v2.5 Fast enough255.191,076 ±11$40 to $50
10 = Soniox TTS Real-Time v2 Visible delay324.7101,180 ±11$13 to $14.81
14 Cartesia Sonic 3.6 Visible delay374.491,292 ±14$39.20 to $50
15 Cartesia Sonic 3.5 Visible delay390.5151,190 ±11$39.20 to $50
18 SpaceXAI Grok TTS Visible delay473.9111,155 ±13$15
19 Fish Audio S2.1 Pro Visible delay533.021,184 ±13$15
not ranked Alibaba Qwen-Audio-3.0-TTS-Plus n/an/a81,269 ±13$19.30
not ranked Boson AI Higgs Audio V3 TTS n/an/a91,045 ±13$15
not ranked BreezeBlue Breeze TTS 2 n/an/a11,227 ±13$34
not ranked Cartesia Sonic 3 n/an/a91,054 ±11$49 to $50
not ranked Coqui XTTS v2 n/an/a7913 ±15$40.40
not ranked ElevenLabs Eleven v3 n/an/a101,184 ±10$100
not ranked ElevenLabs Eleven v3 Conversational n/an/a131,213 ±12$50
not ranked ElevenLabs Eleven v4 n/an/a91,333 ±16$80
not ranked ElevenLabs Multilingual v2 n/an/a81,093 ±10$100
not ranked ElevenLabs Turbo v2.5 n/an/a91,093 ±10$50
not ranked Fish Audio Drama 3 Preview Fast enough209.321,245 ±14$15
not ranked Fish Audio OpenAudio S1 n/an/a71,084 ±11$15
not ranked Fish Audio OpenAudio S1 Mini n/an/a71,043 ±20$15
not ranked Fish Audio S2 Pro n/an/a91,108 ±12$15
not ranked Google Gemini 3.1 Flash TTS n/an/a91,212 ±11$18.30 to $33.30
not ranked Google Gemini 3.8 Flash TTS n/an/a91,280 ±14$15 to $16.50
not ranked Google Gemini 3.8 Flash-Lite TTS n/an/a91,256 ±13$10 to $11
not ranked MiniMax Speech 2.6 HD n/an/a41,126 ±11$100
not ranked MiniMax Speech 2.8 HD n/an/a111,174 ±10$100
not ranked MiniMax Speech 2.8 Turbo n/an/a91,153 ±10$60
not ranked OpenAI GPT-4o mini TTS n/an/a9n/a$20

Frequently asked questions

Which TTS APIs support Chinese?
30 models were tested in Chinese. Starting with the best-ranked, Inworld Realtime TTS-2 Flash, Inworld Realtime TTS-2, ElevenLabs Eleven v4 Turbo, and ElevenLabs Flash v2.5. Being tested in a language is evidence of support, not a quality score.
Is TTS latency measured in Chinese?
No. Coval measures latency in English from the US East, so the latency columns show how fast each model starts in general, not in Chinese.
Which Chinese TTS API is fastest?
Among models tested in Chinese that are also ranked, Inworld Realtime TTS-2 Flash is first, with a P95 of 100.9 ms.
How many of the 30 models are ranked?
9 of the 30 are in the multilingual speed ranking; the others are not tagged multilingual by Coval or are not measured by it.