Chinese
TTS Voices
Chinese text-to-speech voices with accurate tonal precision
From text to talk.
Pick your path.
Call our TTS & STT endpoints directly, wire voice into LiveKit rooms with one plug-in, or spin up an AI assistant on a real phone number.
TTS & STT Endpoints
Production-grade streaming and batch TTS/STT. Low latency, 50+ languages, customizable voices, and SDKs for Node/Python/Browser.
- ›Streaming for live apps
- ›Multi-speaker diarization & punctuation
- ›SDKs, code samples, and latency benchmarks
Sends text to the TTS endpoint and saves the synthesized audio as an MP3 file.
LiveKit Plug-in
Plug our real-time speech pipeline into LiveKit rooms — transcribe live sessions, synthesize responses and stream audio back into the room.
- ›One-line install, example room demo
- ›WebRTC + server bridge patterns
- ›Works in browser & mobile
Connects to a LiveKit room and attaches real-time TTS/STT — transcribes audio in, synthesizes audio out.
AI-Assistants (Phone)
Deploy a phone-number based AI assistant in minutes — inbound/outbound calls, IVR, call recording, and DTMF support.
- ›Purchase & map a phone number
- ›Templates: Support Bot, Sales Assistant, Reminder Bot
- ›PSTN reliability & compliance tools
Creates an AI assistant bound to a phone number with inbound call handling, recording, and DTMF support.
Spanish voices
294TTS voicesEspañol
French voices
98TTS voicesFrançais
German voices
82TTS voicesDeutsch
Indonesian voices
31TTS voicesBahasa Indonesia
Italian voices
51TTS voicesItaliano
Japanese voices
85TTS voices日本語
Korean voices
171TTS voices한국어
Portuguese voices
277TTS voicesPortuguês
Russian voices
34TTS voicesРусский
Chinese voices
189TTS voices中文
Chinese phonology and prosody
Pitch that carries the dictionary
Mandarin is a tone language[1]: nearly every syllable carries one of four fixed pitch contours: high, rising, dipping, or falling: and changing the contour changes the word. The syllable "ma" means mother (mā), hemp (má), horse (mǎ), or scold (mà). English uses pitch to signal stress and focus[2]; Mandarin pitch is built into the word itself[3]. A TTS system that misshapes a single tone doesn't sound unnatural: it says the wrong word. Producing accurate Mandarin requires inference that resolves tone at the syllable level, with no inter-provider routing degrading the pitch signal.
Evenly chopped, not bouncy
English is stress-timed[1]: stressed syllables land at regular intervals while unstressed syllables compress between them, creating a strong-weak-weak bounce. Mandarin is syllable-timed[2]: syllables stay closer to equal in duration with far less reduction, producing what sounds like a row of similarly sized beats[3] carrying different pitch shapes. A voice engine trained on English stress-timing will squeeze and stretch syllables that should stay even. Getting this right requires models built for Mandarin rhythm, running co-located with the audio pipeline.
Sentence melody on a tonal tightrope
English intonation is relatively free: pitch accents move around a sentence to mark focus or signal questions[1] without changing word identity. Mandarin intonation must ride on top of lexical tones[2], using post-focus pitch compression to convey emphasis while keeping each syllable's tone intact. The same discourse function: focus, question, statement: is realized through different prosodic strategies[3] than English uses. Imposing English-style rising question contours onto Mandarin warps lexical tones into wrong words. This two-layer pitch system demands synthesis infrastructure where tone and intonation are resolved together, not split across providers.