TikTok Ads Transcript Scraper — Creative Center Top Ads hooks & transcripts

Turn TikTok Creative Center Top Ads into structured transcripts — full text, the first-3-seconds hook, and CTR/brand metadata in one JSON row per ad. Chain this TikTok ads transcript scraper straight after any Creative Center / Top Ads scraper run (dataset ID, pasted rows, or video URLs), or pass Top Ads material IDs and it fetches a fresh video link itself. You are charged only when a transcript is delivered — expired links, music-only creatives, and ads that dropped out of Top Ads carry no result fee.

You give it	You get back
Ad video URLs (videoUrl / videoUrl720p / video_url from any Creative Center scraper)	transcript — full speech-to-text, any length
…or your scraper run's dataset ID (one-click chaining)	hook3s — what's said in the first 3 seconds
…or Top Ads material IDs / ad page links you copied browsing Creative Center (fresh links fetched for you)	brandName, adTitle, ctr, cost, likes, industryKey, objectiveKey, materialId
…or pasted dataset rows (any scraper's shape)	language, durationSeconds, timestamped segments, urlExpiresAt

Output

One JSON row per ad. Delivered rows carry the transcript and charged: true; anything that could not be delivered carries a status saying why, and charged: false. Real rows from the live example run below (long text trimmed with … for the page):

status	charged	materialId	ctr	likes	hook3s	transcript	language	durationSeconds
transcribed	true	7662608412568305665	0.97	1510	Mystic Skill Enlightenment. Unlock new Mystic Skill Motive in the hands.	Mystic Skill Enlightenment. Unlock new Mystic Skill Motive in the hands. Ever seen someone kick a giant battle, and you're just standing there wondering…	English	110.83
no_audio	false	7668529233501913089	0.9	0	null	null	null	14.72
transcribed	true	7662036896940326919	0.27	7520	Hi, how are you? Good, how are you? Good, thank you. Just table for one? Yeah, absolutely.	Hi, how are you? Good, how are you? Good, thank you. Just table for one? Yeah, absolutely. Perfect. Just got off work. I'm so exhausted…	English	175.74

The middle row is a music-only creative: it is delivered with its Top Ads metadata and not charged.

Full row, as JSON:

{
  "status": "transcribed",
  "charged": true,
  "materialId": "7662340784365568017",
  "brandName": "",
  "adTitle": "Why Promote Stands Out Here is why Promote stands out! Target the audience based on your goals...",
  "ctr": 0.39,
  "cost": 0,
  "likes": 19,
  "industryKey": "label_20112000000",
  "objectiveKey": "campaign_objective_reach",
  "hook3s": "Looking for more ways to support your growth on TikTok?",
  "transcript": "Looking for more ways to support your growth on TikTok? Here's why Promote stands out. Promote is TikTok's in-app advertising tool to grow account performance...",
  "language": "English",
  "durationSeconds": 44.2,
  "urlExpiresAt": "2026-08-21T20:14:56.000Z",
  "segments": [{ "start": 0, "end": 2.96, "text": "Looking for more ways to support your growth on TikTok?" }],
  "chargeEvents": { "transcript": 1, "surchargeMinutes": 0 }
}

Browse the full live example dataset → — real output from a verified run, including the free uncharged rows.

Every field is always present (explicit null over silent omission). Non-delivered rows carry status + statusReason instead — unavailable_expired (dead link, expiry time stated), no_audio (music-only or voice-less creative), no_audio_stream (the file has no audio track), no_onscreen_text (a silent ad with no readable on-screen copy either — see below), no_creative_found (no video link or material ID in the row, or the ad left Top Ads), input_error (guidance on what to pass instead), asr_unavailable (our speech service refused us mid-run — please re-run), or failed_* (also listed in the run's ERRORS record). None of them are charged.

Silent ads: read the on-screen text instead (optional)

A good share of Top Ads carry no speech at all — music-only creatives and kinetic-typography ads whose whole message is on the screen. By default those ship as free no_audio / no_audio_stream rows and every run tells you how many it saw.

Turn on Read on-screen text on silent video ads (includeOnScreenText, off by default) and each silent ad's own keyframes are read instead:

{
  "status": "onscreen_text_extracted",
  "charged": true,
  "materialId": "7668529233501913089",
  "onScreenText": {
    "headline": "Let AI build your Shopify loyalty program in minutes",
    "body": "No coding, no headaches. Just plug in and start rewarding.",
    "cta": "START FOR FREE",
    "rawText": "Let AI build your Shopify loyalty program in minutes\nNo coding, no headaches…"
  },
  "chargeEvents": { "transcript": 1, "surchargeMinutes": 0 }
}

• A delivered onScreenText row is charged like a transcript — same event, same price, never the long-video surcharge.
• Ads with no readable on-screen copy stay uncharged either way: b-roll whose only visible text is a logo or a placard comes back as no_onscreen_text, not as a charged row of noise.
• The frames come from the video this actor already downloaded, so nothing is fetched twice.
• brandName always comes from TikTok's own Top Ads metadata, never from reading a video frame.

For AI agents & LLMs — copy this block

Actor:   steadyfetch/tiktok-ads-transcript-scraper  (Apify Store)
Does:    TikTok Creative Center Top Ads -> full transcript + first-3-seconds hook +
         CTR/brand metadata. One JSON row per ad. No TikTok login, cookies or API key.
Run:     POST https://api.apify.com/v2/acts/steadyfetch~tiktok-ads-transcript-scraper/run-sync-get-dataset-items?token=APIFY_TOKEN
Input    (any ONE of the first four fields is enough):
  {
    "materialIds": ["7662340784365568017"],   // Top Ads material ids; a fresh video link is fetched
    "videoUrls": ["https://ads.tiktok.com/business/creativecenter/topads/7662340784365568017"],
                                              // ad page links or *.tiktokcdn.com video links
    "datasetId": "<dataset id of any Creative Center / Top Ads scraper run>",
    "datasetItems": [ { /* pasted scraper rows, any shape */ } ],
    "includeOnScreenText": false,             // optional: read silent ads' on-screen text
    "maxAds": 1000                            // optional cap
  }
Output   one row per ad: status, charged, materialId, brandName, adTitle, ctr, cost,
         likes, industryKey, objectiveKey, hook3s, transcript, language,
         durationSeconds, urlExpiresAt, segments[]
         with includeOnScreenText: silent ads add onScreenText{headline,body,cta,rawText}
Pricing  pay per event, platform usage INCLUDED, no start fee:
  tiktok-ad-transcript   $0.020 -> $0.008 per delivered transcript (falls with your Apify plan tier)
                         also charges a delivered onScreenText row (includeOnScreenText)
  long-video-surcharge   $0.005 per started minute past the first 3 minutes
  Charged only on delivery — expired links, failures, and silent ads with no
  readable text cost $0.
  Cap any run with maxTotalChargeUsd (minimum $0.05).
MCP:     reachable through Apify's MCP server; ask for "TikTok ad transcripts".

Why this TikTok ads transcript scraper?

• Charged only on delivery. A transcript either lands in your dataset or the row carries no result fee. Expired link → uncharged row that says so, with the exact expiry time. Music-only ad → uncharged row that says so. No start fee, no per-run fee.
• Built for TikTok's 6-hour links. Creative Center video links are signed and expire about 6 hours after your scraper minted them (not the 24–48 hours some listings claim). This actor decodes the expiry before downloading, skips dead links at zero cost, and — if your rows carry the ad's material ID — fetches a fresh link from the Top Ads detail page and transcribes anyway. Old datasets become usable again.
• Any video length. Top Ads run 5 seconds to 5 minutes; all of it transcribes. Videos past 3 minutes add a small per-started-minute surcharge instead of failing or being cut.
• Hook as data. The first-3s hook is the line media buyers study; you get it as its own field, plus full segment timestamps for anything deeper.
• Works with any scraper's output. The input deep-scans your rows for the TikTok CDN video link and the ad metadata — no field mapping, no glue code.
• Priced for bulk. The first 3 minutes of every ad are included, and the per-transcript price falls with volume — see the pricing tab.

This actor may fail when the platform changes things — failed items are never charged.

How to transcribe TikTok Top Ads (no code)

1. Run any TikTok Creative Center / Top Ads scraper (see Works with below) — pick your region, industry, and period as usual.
2. Open this actor and paste the scraper run's dataset ID into Dataset ID (or use Apify's Connect Actor integration so it chains automatically the moment the scraper finishes — the best way to beat the 6-hour link expiry).
3. Click Start. Each ad comes back as one JSON row: transcript + hook + CTR/brand metadata.
4. Export as JSON/CSV, or read it via API.

Prefer URLs? Paste the videoUrl / videoUrl720p / video_url values into Ad video URLs. Browsing Creative Center and found an ad you want transcribed? Copy its page link (ads.tiktok.com/business/creativecenter/topads/<id>, with or without the trailing /pc/en and tracking parameters) and paste it into either field — same for material IDs. The actor fetches a fresh video link for each.

Why do TikTok Creative Center video links expire?

TikTok's video CDN signs every link with an expiry timestamp — about 6 hours from when the page (or your scraper) minted it. After that the link is dead. This actor handles it honestly:

• A link that is already expired is refused before any download — the row comes back as uncharged unavailable_expired with urlExpiresAt showing exactly when it died.
• If the same row carries the ad's material ID (most scrapers include it), the actor fetches a fresh link from the ad's public Top Ads detail page and transcribes anyway (best effort — if the ad has since dropped out of Top Ads, you get an uncharged no_creative_found row).
• If most of a run's links are expired, the run summary says so plainly and suggests the fix: re-run your scraper, or pass material IDs.

Transcribe soon after scraping, or chain automatically — Connect Actor on the scraper run hands this actor the dataset while every link is still fresh.

How much does it cost to transcribe TikTok ads?

One result = one ad's full transcript payload (transcript + hook + metadata + segments). The first 3 minutes of each video are included; longer videos add a small per-started-minute surcharge. No start fee, no subscription, no third-party API key to buy.

Job	Approx. cost
50 Top Ads from one industry	≈ $1.00
500 ads (a serious creative teardown)	≈ $10
Apify free plan ($5 credit)	≈ 250 ad transcripts

Platform usage (compute + transfer) is included in the event price — the per-transcript price you see is the whole price. Expired, music-only, or dropped ads carry no result fee.

Use it via API, MCP, and integrations

• API: standard Apify run API — POST .../acts/steadyfetch~tiktok-ads-transcript-scraper/runs with { "datasetId": "..." } or { "materialIds": ["..."] }. Python/Node clients work as with any actor.
• MCP: works from Claude, Cursor, and any MCP client through Apify's MCP server — ask for "TikTok ad transcripts" with a dataset ID, material IDs, or URLs.
• n8n: a free ready-made template, no community nodes needed — TikTok Top Ads transcripts → (Creative Center industry/region → hooks & transcripts → spreadsheet-shaped rows, 3-minute setup).
• Make / Zapier: call it as a regular Apify actor step right after your Creative Center scraper node — pass the scraper's default dataset ID.

Works with — Creative Center scrapers you can chain from

This actor reads the video link value in each row (any https://…tiktokcdn…/… video link), not a fixed field name, so it works with any Creative Center scraper that exports direct video links — and it picks up the material ID for the fresh-link fallback. Verified output shapes on this store:

Scraper	Where the video link lives	Material ID field
azzouzana/tiktok-creative-center-top-ads-scraper	videoUrl720p / videoUrl540p / videoUrl360p	adId (also detailUrl)
doliz/tiktok-creative-center-scraper	video_info.video_url.720p	id
datapeak/tiktok-creative-center	videoUrl	adId (also detailsUrl)
beyondops/tiktok-ad-library-scraper	videoUrlHd / videoUrl	adId
lexis-solutions/tiktok-top-ads-scraper	videoUrls.720p	id
dltik/tiktok-creative-center	video_url_720p	ad_id (also detail_url)

How to chain: run the scraper → copy the run's dataset ID into this actor's Dataset ID field, or use Apify's Connect Actor integration to pass it automatically → Start. Using a scraper that isn't listed? If its rows contain the TikTok CDN video link or the material ID, it works the same way — and if a row has neither, it comes back as an uncharged no_creative_found.

FAQ

What happens to expired links? They are detected before any download and returned as uncharged unavailable_expired rows with the expiry time. If the row also has a material ID, the actor fetches a fresh link first and only reports expiry when that fails too.

Music-only ads, or ads without a voice-over? Speech-recognition models hallucinate filler on music. The actor detects that (in any language) and returns an honest, uncharged no_audio row instead of selling you a fake transcript. Expect a fair share of these on TikTok — many top ads are music plus on-screen text.

Do I need TikTok cookies or an API key? No. The actor only downloads public ad videos from TikTok's CDN and reads public Top Ads detail pages. No login, no cookies, no third-party ASR key.

Can I paste the Creative Center listing page? No — that page is a list of many ads, not a single ad. Run a Creative Center scraper over it first and chain the result here (or open one ad and paste its page link / material ID). A listing URL comes back as an uncharged input_error row with this guidance.

Material IDs — what if I mistype one? Material IDs are long numbers, so a typo can look like a valid ID. Those come back as uncharged no_creative_found rows ("no longer listed in Top Ads"), not as input errors — check the ID against your scraper's row if you see one unexpectedly.

What happens on a failed download? Up to 3 attempts, then an uncharged failed_download row plus an entry in the run's ERRORS record. You pay no result fee for anything that didn't deliver.

Languages? Whisper-class multilingual ASR — English, Spanish, Japanese, Arabic, Indonesian, and ~90 more transcribe out of the box; the detected language ships on every row.

Why not scrape Top Ads here too? Single-purpose by design: your scraper (whichever you like — region, industry, keyword filters included) finds the ads; this actor turns them into transcript data, reliably. One job, done honestly.

Is this an official TikTok product? No. This is an independent actor from steadyfetch — unofficial, and not affiliated with, endorsed by, or sponsored by ByteDance Ltd. or TikTok. "TikTok" and "TikTok Creative Center" are trademarks of their respective owners, used here only to name the public Top Ads surface this actor reads.

Steadyfetch ad-intelligence suite

One actor per ad surface, one job each, the same contract everywhere: all-inclusive pay-per-event, no start fee, charged only on delivery — and the same row shape, so a workflow built for one platform ports to the next.

Ad creatives → transcripts & on-creative text

Ad surface	Actor	Free n8n template
Facebook & Instagram Ad Library	Facebook Ad Library Transcript Scraper — Video & Image Ad Text	Facebook ad transcripts → · Competitor ad teardown →
Google Ads Transparency Center	Google Ads Video Transcripts & Hooks (Ads Transparency Center)	Google Ads video transcripts →
TikTok Creative Center Top Ads	this actor	TikTok Top Ads transcripts →
LinkedIn Ad Library	LinkedIn Ads Transcript Scraper — Video Hooks & CTAs	LinkedIn Ad Library transcripts →
Instagram Reels (organic posts)	Instagram Reels to Text · @handle	Instagram Reel transcripts →

Ad creatives → text (image & text ads, read by OCR)

Ad surface	Actor
Google Ads Transparency Center	Google Ads Creative Text & OCR (Transparency Center)

All templates in one place: github.com/steadyfetch/n8n-templates — free, no community nodes needed.

The rest of the steadyfetch shelf — same contract everywhere: all-inclusive pay per event, no start fee, charged only on delivery.

Family	Actors
Ad creative intelligence	Facebook · Google Ads video · TikTok · LinkedIn · Google Ads text & OCR
Trends & keywords	Google Trends · Trends Now · Breakout keywords · Autocomplete keywords · Keyword volume & CPC · Social trends
YouTube transcripts	YouTube videos · YouTube channels
Instagram	Reel transcripts · Profile posts
Jobs	Indeed · Career sites by domain · Glassdoor · Multi-board · Google Jobs
Amazon	Products · Search · Bestsellers · Sellers
Any media file	Speech to Text · any link or file

Unlinked names are publishing shortly on the same account — search steadyfetch on Apify Store.

Feedback & support

Found an issue? Open it on the Issues tab — we respond within one business day. Feature requests welcome.