Douyin Transcript Scraper avatar

Douyin Transcript Scraper

Pricing

from $7.00 / 1,000 audio minute transcribeds

Go to Apify Store
Douyin Transcript Scraper

Douyin Transcript Scraper

Transcribe Douyin (抖音) videos to Chinese text without login. Paste video links, share links or creators and get timestamped speech-to-text with punctuation (SenseVoice), plus likes, author and music. Videos without speech are not charged.

Pricing

from $7.00 / 1,000 audio minute transcribeds

Rating

0.0

(0)

Developer

axly

axly

Maintained by Community

Actor stats

0

Bookmarked

2

Total users

1

Monthly active users

3 days ago

Last modified

Share

Douyin Transcript Scraper (抖音视频转文字)

Turn Douyin videos into text. Paste video links, share links or whole creators and get a timestamped Chinese transcript of what is said in each video, with punctuation, plus likes, comments, shares, author and music. Speech recognition runs on SenseVoice, a model built for Mandarin (and Cantonese, English, Japanese, Korean), so the text comes out in simplified Chinese with punctuation instead of an unpunctuated character stream. No login, cookies or API keys.

Who it is for

  • Marketers, agencies and cross-border sellers: study the scripts and hooks of top Douyin creators in your niche.
  • Researchers and analysts: build spoken-content datasets for NLP, trend and narrative research.
  • AI and RAG pipelines: feed Douyin speech into LLMs, translators and search indexes as clean text.
  • Content teams: repurpose videos into articles, subtitles and translations.

What you get

FieldExampleNotes
url, videoIdhttps://www.douyin.com/video/7696934197226786058
authorNickname, authorSecUid心影说
description, hashtags法律迟迟不肯给出公道,八旬老兵该如何讨回尊严?the post text
createdAtUTC
durationSeconds664.7
likeCount, commentCount, shareCount, collectCount163, 2, 10, 21
transcript[{"start": 0.0, "end": 9.8, "text": "老人架起狙击步枪,枪口牢牢锁定了对面的三个败类,…"}]timed segments, or one string
hookthe opening sentence (segments starting in the first 3 seconds)
languagezhdetected per video (zh, yue, en, ja, ko)
characterCount, hasSpeech, audioSeconds
musicTitle, musicAuthor, audioSource@心影说创作的原声, original_sound_mp3
error, errorReasonno_speechset when a video could not be transcribed

Videos without speech (music-only clips), image posts (图文) and videos that fail are listed with an error code so you can see what happened to every input, and you are not charged for them.

How to use it

  1. Paste Douyin videos: douyin.com/video/… links, share links (v.douyin.com/…) or numeric video IDs.
  2. Or add creators: profile links (douyin.com/user/MS4wLjAB…). The actor transcribes their newest videos, up to Max videos per creator (Douyin shows logged-out visitors about the 40 newest videos of a creator).
  3. Run, then export the dataset as JSON, CSV or Excel, or use the API.

Input options

OptionDefaultWhat it does
videoUrls–video links, share links or IDs
profileUrls–creator profile links
maxVideosPerProfile20newest videos per creator
maxVideos50cap for the run
maxAudioMinutes30longer videos are skipped (error: too_long)
outputFormatsegmentssegments, text or both
proxyConfigurationResidentialused only for Douyin's small API requests; audio is downloaded directly

Example input

{
"videoUrls": ["https://www.douyin.com/video/7696934197226786058"],
"profileUrls": ["https://www.douyin.com/user/MS4wLjABAAAAOzRPcYBhm0xO0mXI5UGjHc2udHifSV-6xuI5mpo_UNKFN6qLVzfWb4QMT8BTOk2n"],
"maxVideosPerProfile": 5,
"outputFormat": "both"
}

Scheduling and integrations

Schedule the actor with your creators' profile links to transcribe their new videos regularly, and send the dataset to Google Sheets, Slack, Make, Zapier or a webhook. A run interrupted mid-way resumes where it stopped when started again with the same input; videos that failed are retried on the next run.

Use with AI assistants (MCP)

Apify's MCP server exposes this actor as a tool, so Claude, ChatGPT, Cursor and other MCP clients can call it directly: "Transcribe the 10 newest videos of this Douyin creator and summarize them in English".

FAQ

Do I need a Douyin account? No. Only public videos are read, logged out.

How accurate is it? SenseVoice handles clear Mandarin narration well and adds punctuation. Background music, dialects, heavy slang or several people talking at once lower accuracy. Song lyrics are transcribed like speech.

Does Douyin provide subtitles? Not in its public web data, so every transcript is generated from the audio.

Which languages? Mandarin, Cantonese, English, Japanese and Korean are detected automatically.

How many videos per creator? Douyin shows logged-out visitors roughly the 40 newest videos of a creator; older videos can be transcribed by pasting their links.

Is this legal? The actor reads publicly available videos without logging in. You are responsible for how you use the data, in particular personal data and Douyin's terms.

Something failed? Rows carry an error code and reason, and failed videos are retried when you re-run the same input. Open an issue on the actor page if it keeps happening.