Free tool
Chinese Text to Speech
Paste Chinese text and hear it spoken. Words highlight in time with the audio, pinyin sits above every character, and you can download the MP3 or share a link to a playback page.
About this tool
Reading Chinese and hearing Chinese are close to separate skills. A learner who can work through a paragraph on the page often cannot follow the same paragraph at conversational speed, because the characters give no clue about where one word ends and the next begins, and tones only exist in sound.
This tool closes that gap. It reads your text aloud with natural Mandarin pronunciation, and as it plays, each word lights up in time with the voice — so you can see the segmentation you cannot hear yet. Pinyin sits above every character, so you can check a reading without leaving the passage.
You can scrub through the audio, replay a difficult clause, download the file for offline shadowing practice, or copy a link that opens a dedicated playback page with the same synchronised text — useful for sending a pronunciation model to a student or a colleague.
What it does
Karaoke word highlighting
Words highlight as they are spoken, driven by real timestamps from the audio rather than a guessed average pace. This is the fastest way to internalise where Chinese word boundaries actually fall.
Pinyin above every character
Every character carries its pinyin directly above it, so you can read and listen at the same time without looking anything up.
Full playback control
A draggable progress bar, play and pause, and sentence-level skip so you can loop a clause until it sticks.
Download the audio
Save the generated speech as an audio file for offline listening, shadowing drills or a lesson plan.
Shareable playback page
Copy a link that opens a standalone page with the text, the audio and the same synchronised highlighting — no account required for whoever opens it.
Natural Mandarin voice
Standard Mandarin pronunciation with correct tones and natural sentence prosody, rather than the flat character-by-character reading of older systems.
How to use it
- 1Paste your Chinese textA sentence, a dialogue, a paragraph from a textbook — anything you want to hear read aloud.
- 2Generate the audioThe text is segmented into words and read aloud. Generation takes a few seconds for a normal paragraph.
- 3Follow alongPress play and watch the highlighting track the voice, with the pinyin above each character to check any reading you are unsure of.
- 4Download or shareSave the audio for offline practice, or copy the playback link to send the whole synchronised reading to someone else.
What people use it for
- Shadowing practice — listening and repeating in real time to fix pronunciation and rhythm
- Checking how a sentence you wrote actually sounds before you say it
- Turning a textbook passage into listening material when the book has no audio
- Sending a pronunciation model to a student, with the text synchronised
- Making Chinese content accessible for people who find reading characters difficult
Frequently asked questions
Is the audio natural-sounding?
It uses a neural Mandarin voice, so tones, sentence intonation and pausing follow natural speech rather than reading each character in isolation. It is close enough to a native reading to be useful as a pronunciation model for learners at any level.
How does the karaoke highlighting stay in sync?
The speech engine returns per-word timestamps alongside the audio, and those timestamps are matched against the segmented text. The highlight is driven by the audio element's actual playback position, so it stays accurate even when you scrub, pause or replay.
Can I download the audio file?
Yes. The download button saves the generated speech as an audio file you can keep, put on a phone, or drop into a lesson.
What does the share link do?
It opens a standalone playback page showing your text with the same word-level highlighting and pinyin above the characters, plus the player. Anyone with the link can open it — no sign-in, no app.
Which Chinese does it speak?
Standard Mandarin (Putonghua), the variety taught in courses and used in broadcasting. Both simplified and traditional input are read correctly, since the difference is in the writing system rather than the pronunciation.
Does it handle polyphonic characters correctly?
In the great majority of cases, yes — the engine works from word-level segmentation, so 银行 is read yínháng and 行为 is read xíngwéi. Rare proper nouns are the usual exception, so check names if they matter.
How long can the text be?
Comfortably a paragraph or two. Very long documents are better split into sections, which is also better practice — a passage you can replay ten times is worth more than one you listen to once.
Guides and further reading
In-depth articles covering the background behind this tool.
Other Chinese tools
Last reviewed