Start with 45 free minutes of callsTry for free

Text-to-Speech · Multilingual

Chinese (Mandarin)

Language

Request a new language

Tell us which language you want added to this benchmark.

Benchmarks

X
Y
200ms400ms600ms800ms1400150016001700Time to first audio lower is betterNaturalness higher is betterInworld tts-2ElevenLabs eleven_v3_conversationalGemini 3.1-flash-tts-previewCartesia sonic-3.5MiniMax speech-2.6-hdOpenAI gpt-4o-mini-ttsspeko.ai/benchmarks

Naturalness

higher is better

1500 is this language's own field — not comparable across languages.

field averagetied with bestOpenAIgpt-4o-mini-ttsTry →1365MiniMaxMiniMaxspeech-2.6-hdTry →1467AzureAzureAI Speech (native)1552Cartesiasonic-3.5Try →1564GeminiGemini3.1-flash-tts-preview1649ElevenLabseleven_v3_conversationalTry →1675InworldInworldtts-2Try →1688
1200140016001800

Time to first audio

lower is better

The pause the caller hears before the agent starts speaking.

InworldInworldtts-2Try →121ms · 133ms
Cartesiasonic-3.5Try →137ms · 193ms
MiniMaxMiniMaxspeech-2.6-hdTry →340ms · 465ms
OpenAIgpt-4o-mini-ttsTry →602ms · 944ms
GeminiGemini3.1-flash-tts-preview634ms · 823ms
ElevenLabseleven_v3_conversationalTry →890ms · 1.13s
0ms300ms600ms900ms

p50 p90

Robustness

higher is better

Whether the voice says the right words for a number, date, currency amount or operator.

SpeechifyTry →1.00
Gemini0.75
Try →0.75
MiniMaxTry →0.75
InworldTry →0.50
Try →0.25
Try →0.00
0.000.250.500.751.00

Cost

lower is better

$ per 1M characters · vendor list price, the same in every language