Chinese subtitles generated from the video speech
TOPIC HUB
Chinese Video Subtitles
Put readable Chinese subtitles on a video you already have, with pinyin aligned to every character and English available when you want it.
Try the Toolchinese video subtitles
OUTPUT PREVIEW
你吃饭了吗？
Have you eaten?
QUICK ANSWER
Chinese video subtitles are the timed Chinese text of what is spoken in the video, shown over the picture so it can be read while it plays. On this site the track can carry three layers — Chinese characters, pinyin aligned above each character, and an English translation — and it exports as SRT or ASS. The Chinese comes from the video itself: you upload a file, the speech is recognised into timed Chinese text, and you review it before exporting.
EXPLORE THE CATEGORY
Choose the workflow that matches your video.
Pinyin aligned one-to-one above each character
English hidden by default, available as a layer
HOW IT WORKS
From a Chinese video file to a subtitle file.
- 01
Upload the video from your own device — MP4, MOV or MKV. The workflow starts from a local file rather than a streaming link, so the source has to be something you already hold.
- 02
The Chinese speech is recognised and split into timed segments. Every segment gets a start and an end, and that timing is what turns a transcript into subtitles.
- 03
Review the Chinese line by line in the editor. Clicking a line moves the player to that moment, so a character can be checked against what was actually said instead of guessed from context.
- 04
Switch on the layers you want and export. Chinese is always present; pinyin and English are optional. SRT goes anywhere, ASS carries positioning for a specific frame.
SPECIFICATIONS
The three layers, and what each one is for.
| Layer | What it shows | When to switch it on |
|---|---|---|
| Chinese | The characters that were spoken, timed to the video | Always — this is the subtitle |
| Pinyin | The reading, aligned above each character | When the audience is learning Chinese, or a character is unfamiliar |
| English | A translation of the Chinese line | When the meaning has to be understood, not just heard — and it can stay hidden on the video |
Why read Chinese subtitles instead of English ones
Watching a Chinese video with English subtitles is comfortable and teaches very little. The eyes track the English, the ear receives a stream of sound it cannot segment, and nothing connects the two. Reading the Chinese is harder at first, and it is the only version that builds the link between sound, character and meaning.
Pinyin sits between those two extremes, which is why it belongs on the same track rather than in a separate document. Above each character it gives the reading at the moment that character is heard, so a word that would otherwise be skipped becomes recognisable — and the next time it appears it is read without the pinyin.
SRT or ASS: which subtitle file to export
SRT carries text and timing and nothing else. That is exactly why it is safe: practically every player, editor and upload form accepts it. If the file is going somewhere you do not control, SRT is the format that cannot be rejected for being unsupported.
ASS carries the same text and timing plus positioning, font, size and margin information. It is the right export when the subtitles have to sit inside a specific frame — a vertical clip, a square social edit, or a video where the text must avoid a graphic already in the lower third. The tradeoff is that fewer tools read ASS directly, so it usually ends up burned into a render rather than shipped as a sidecar file.
RELATED TOOLS
Explore related pages.
WHEN THIS IS NOT WHAT YOU NEED
When this is not the route you want.
- If you have no Chinese video yet, there is nothing to subtitle. The workflow reads the speech in a file you already hold; it does not fetch video from a platform.
- If you need the video dubbed rather than subtitled — voice replaced, lip sync adjusted — that is a different class of tool. This produces a text track with timing.
- If an accurate subtitle track already exists for the file, edit that instead. Re-running recognition over audio that already has good captions adds errors rather than removing them.
- If the recording is mostly music, singing, or several people talking over one another, expect to spend most of the time in the transcript editor. Recognition assumes one clear speaker at a time.
The honest framing: recognition produces the draft, review produces the subtitles. Judge the result after the review pass, not before it.
QUESTIONS
Questions about Chinese video subtitles.
What are Chinese video subtitles?
The timed Chinese text of what is spoken in a video, displayed over the picture. On this site the track can include three layers — Chinese characters, pinyin aligned above each character, and an English translation — and it exports as SRT or ASS.
Do I need a Chinese video, or an English one?
A Chinese one. The video supplies the speech that becomes the subtitles, so the workflow reads Chinese audio: spoken Mandarin in a local file. It is not a tool for turning English video into Chinese text.
Can the track show pinyin and English at the same time as Chinese?
Each layer is controlled separately. Chinese is always present because it is the subtitle; pinyin and English can each be switched on or off. English can stay hidden on the video while remaining available for reference, which is how one file serves both viewing and study.
Should I export SRT or ASS?
SRT if the file goes to a player, editor or upload field you do not control — it carries text and timing and is accepted nearly everywhere. ASS when the subtitles need positioning, font or margins for a particular frame, accepting that fewer tools read ASS directly.
How finished are the subtitles when they are first generated?
Close on clear speech, less close on names and unusual terms. Character-level recognition is strong, and the failures concentrate where context is needed to pick the right character. The editor exists for that pass, and clicking a line replays the exact moment it came from.
Is there a free way to try it?
There is a free allowance in minutes — enough to run a short clip end to end and judge how your own audio behaves, which is the only test that tells you anything useful about your recording.