Facebook Reels Scraper + Transcripts

Get public Facebook Reels with full metadata and what is said in them. Paste reel links, or just page names like cnn or facebook.com/bbcnews, and get one clean row per reel: caption, plays, reactions, comments, shares, owner, video and thumbnail URLs, plus a spoken-word transcript with timed segments.

No login, no cookies, no browser and no residential proxy needed.

Quick start (first run in a few minutes)

1. Click Start. The form is prefilled with https://www.facebook.com/bbcnews/reels/ and 3 reels per page, with transcripts on auto.
2. Wait about 1-2 minutes (Whisper runs on the Actor's own CPU).
3. Open the Reels table: 3 rows with metadata and a transcript for every reel that has speech.

Approximate cost of the prefilled run (Free plan): about $0.04 (start $0.02 at the default 4096 MB, 3 reels × $0.0025, and up to 3 Whisper transcripts × $0.004).

What you can do with it

• Content research: see which reels of a page or competitor get the most plays, reactions and shares, and what they actually say.
• Repurposing: turn reels into blog posts, captions, subtitles (SRT) or quotes.
• Brand and news monitoring: track new reels of brands, publishers or creators every day with a schedule.
• AI and NLP datasets: searchable spoken text with language detection for classification, summaries or sentiment.
• Ad and influencer vetting: check what a creator says on camera before a deal.

Input

Field	What it does	Default
startUrls	Reel links (/reel/123…, /watch?v=…, /<page>/videos/…, fb.watch/…, /share/r/…), page URLs (facebook.com/cnn, facebook.com/cnn/reels) or bare handles (cnn, @bbcnews). Mix freely.	—
maxReelsPerPage	Newest reels per page (page URLs/handles only), up to 2000	20
transcriptMode	auto (Facebook captions, then Whisper), native (captions only), whisper (always Whisper), none	auto
whisperModel	base (fast) or small (more accurate, about 2.5× slower)	base
transcriptLanguage	Optional spoken-language hint (en, de, es …); empty = auto-detect per reel	—
maxWhisperTranscripts	Hard cap on Whisper transcripts per run	200
maxWhisperDurationSecs	Skip Whisper for longer reels (metadata is still returned)	180
includeSegments / includeSrt / includeVideoUrls	Timed segments, SRT subtitles, signed video URLs	on / off / on
proxyMode	auto (direct first, Apify datacenter proxy as fallback), proxy, none	auto

A page's latest 50 reels with transcripts

{"startUrls": ["https://www.facebook.com/bbcnews"], "maxReelsPerPage": 50}

Specific reels, metadata only (cheapest)

{"startUrls": ["https://www.facebook.com/reel/2326466588106158"], "transcriptMode": "none"}

Many pages, Facebook captions only (no Whisper charge)

{"startUrls": ["cnn", "natgeo", "gordonramsay"], "maxReelsPerPage": 100, "transcriptMode": "native"}

Output

One row per reel (example, shortened):

{
  "url": "https://www.facebook.com/reel/2326466588106158",
  "reelId": "2326466588106158",
  "sourcePage": "bbcnews",
  "caption": "At least 64 people have been killed …",
  "createdAt": "2026-10-28T15:30:15Z",
  "durationSecs": 19.69,
  "playCount": 980000,
  "reactionCount": 10901,
  "commentCount": 503,
  "shareCount": 412,
  "owner": {"name": "BBC News", "id": "…", "url": "https://www.facebook.com/bbcnews", "isVerified": true},
  "thumbnailUrl": "https://scontent…",
  "videoUrlSd": "https://video…",
  "transcript": "Officials say the high-speed train …",
  "segments": [{"start": 0.0, "end": 4.2, "text": "Officials say the high-speed train …"}],
  "transcriptSource": "whisper",
  "transcriptLanguage": "en",
  "transcriptStatus": "ok",
  "status": "ok"
}

Fields: url, reelId, postId, caption, createdAt, durationSecs, width/height, playCount (page mode), reactionCount, commentCount, shareCount, owner.*, thumbnailUrl, videoUrlSd, videoUrlHd, audioTrack, isOriginalAudio, captionTracks (every caption Facebook offers, original and translated), transcript, segments, srt, transcriptLanguage, transcriptSource (facebook_captions or whisper), transcriptStatus.

The dataset has two views, Reels and Transcripts only. The OUTPUT record has a run summary (per-page counts, transcript statuses, Whisper seconds).

Transcript status

transcriptStatus	Meaning	Whisper charged?
ok	Transcript returned	yes, only if the source is whisper
no_speech	Music, ambience or silence. No made-up text like "Thank you."	no
low_confidence	Song lyrics, chants or very short or unclear speech; text returned but flagged	no
skipped_budget / skipped_duration	Over your Whisper cap or length limit	no
skipped_memory	Run memory below 2048 MB, so Whisper was skipped	no
no_captions	native mode and Facebook has no captions	no
no_audio / error	No audio track, or the transcript failed	no

Only about 1 in 6 reels has Facebook's own captions. This Actor uses those first (exact and included in the reel price) and falls back to built-in Whisper speech-to-text for the rest.

Pricing

Pay per event, with volume discounts on higher Apify plans:

Event	What it is	Free plan	Bronze	Silver	Gold+
apify-actor-start	Once per run, per GB of memory (default 4096 MB = 4 events)	$0.005	$0.005	$0.005	$0.005
reel	One reel row with metadata; a Facebook-caption transcript is included	$0.0025	$0.0022	$0.002	$0.0017
whisper-transcript	Add-on, only when Whisper returns a transcript with status ok	$0.004	$0.0036	$0.0032	$0.0028

• Free: invalid_input, not_found, no_reels, page_error and error rows, duplicates, and every transcript status other than Whisper ok.
• Examples (Free plan, 4096 MB):
  • 100 reels, metadata + Facebook captions (native): $0.02 + $0.25 = $0.27
  • 100 reels with auto transcripts (about 60 need Whisper and have speech): $0.27 + $0.24 = about $0.51
  • 1,000 reels from 20 pages, native: about $2.52
• Want metadata only for many reels? Use transcriptMode: "native" or "none" and 1024 MB to lower the start fee to $0.005.
• Set Max total charge per run in the run options to cap spend. The Actor stops cleanly when the cap is reached.

How it works and limitations

• Page mode reads the page's public reels tab and pages through it, so one request returns many reels with plays, reactions, comments and shares.
• Whisper (faster-whisper base, int8) runs inside the Actor. Use at least 2048 MB; 4096 MB (one full vCPU) is the default and fastest, at about 8 seconds per one-minute reel.
• Only public reels are supported; private, age-gated, login-only or region-locked content is not.
• playCount comes from page listings. Single reel links do not expose it when logged out.
• Video and thumbnail URLs are signed by Facebook and expire after a few hours, so download them soon after the run.
• Some pages have no reels tab; they return a free no_reels row.
• Network: direct connection first, Apify datacenter proxy as fallback. No residential proxy needed.
• Facebook changes its internals often. If a page returns page_error, please open an issue and we will fix it fast.

FAQ

Why did I get 0 results? Check that the page is public and actually has a Reels tab (open facebook.com/<page>/reels in a private browser window). Pages without reels return a no_reels row. Reel links that were deleted or made private return not_found. Neither is charged.

Why does a reel have no transcript? It is probably music or silence (no_speech), or you used native mode and Facebook has no captions (no_captions). Check transcriptStatus.

Which languages work? Whisper auto-detects about 100 languages. Set transcriptLanguage if you know the language, for better accuracy on short reels.

Can I get only the transcripts? Yes: open the Transcripts only dataset view, or export with the fields url,transcript.

Do I need a Facebook account or cookies? No. Everything is public data, fetched logged-out.

Can I run it every day? Yes. Create a schedule with your pages and a small maxReelsPerPage; duplicates inside a run are removed and never charged.

Is this legal? It only reads publicly available data. You are responsible for how you use it, including GDPR and Facebook's terms. Do not collect personal data you do not need.

More from LatentLark

• TikTok Transcript Scraper: transcripts of TikTok videos.
• TikTok Ads Library Scraper: TikTok EU Ad Library and Top Ads.
• Google Trends Compare: unlimited keywords on one Google Trends scale.
• Google Ads Transparency & LinkedIn Ad Library Scraper: your competitors' ads.
• Google Sheets Export & Import (No OAuth): send any dataset to Google Sheets.