How to transcribe the reels of an Instagram account, with SRT files

What you can and cannot read on an account

Without signing in, Instagram shows the 12 latest posts of a public account, pinned posts first. That is what the scraper reads; no tool that stays signed out sees further back. Older reels can still be transcribed one by one from their links. Meta's own API does not help here: it serves professional accounts their own media, and no transcript at all (details in our guide on Instagram transcripts).

A carousel with several videos gives one record per video, numbered by carouselIndex. A reel found both in an account and as a link is transcribed and charged once.

A text and a subtitle file per reel, in Python

import csv
import pathlib
from apify_client import ApifyClient

client = ApifyClient("<YOUR_APIFY_TOKEN>")
run = client.actor("quarra/instagram-transcript-scraper").call(run_input={
    "urls": ["@bbcnews"],
    "maxVideosPerAccount": 6,
    "includeSrt": True,
})

folder = pathlib.Path("bbcnews")
folder.mkdir(exist_ok=True)
reels = list(client.dataset(run.default_dataset_id).iterate_items())
reels.sort(key=lambda r: (r.get("position") or 0, r.get("carouselIndex") or 0))  # records arrive as reels finish
with open(folder / "index.csv", "w", newline="", encoding="utf-8") as f:
    w = csv.writer(f)
    w.writerow(["videoId", "publishedAt", "durationSeconds", "language", "wordCount", "likeCount", "status", "caption"])
    for r in reels:
        if r["status"] == "ok":
            name = r["videoId"] + (f"-{r['carouselIndex']}" if r.get("carouselIndex") else "")
            (folder / f"{name}.txt").write_text(r["transcript"], encoding="utf-8")
            (folder / f"{name}.srt").write_text(r["srt"], encoding="utf-8")
        w.writerow([r.get("videoId"), r.get("publishedAt"), r.get("durationSeconds"), r.get("language"),
                    r.get("wordCount"), r.get("likeCount"), r.get("missingReason") or r["status"],
                    (r.get("caption") or "")[:100]])
print(run.status, len(reels), "reels ->", folder.resolve())

Tested with apify-client 3.3 on October 10, 2026, in a 30-second call. It wrote 5 text files, 5 SRT files and this index:

On that last reel, plain Whisper writes “Thank you.”: the phrase it tends to invent on music. A reel without a transcript keeps its row in the index, with its reason and no file. With apify-client 2.x, write run["defaultDatasetId"] instead of run.default_dataset_id.

The start of the first SRT file, ready to edit or to upload as captions:

1
00:00:00,000 --> 00:00:06,000
Somehow a man got into the enclosure of Yorkshire Wildlife Park's land of the tigers

2
00:00:06,000 --> 00:00:10,800
with a 300m long raised walkway above allowing visitors to get close enough

Keep up with an account

Schedule the Actor on Apify with the same input, daily for example, and keep the reels you have not seen yet: the videoId of each record is stable, so a set of the IDs already saved is enough. Each run costs the transcripts of the reels it reads, so set maxVideosPerAccount to what an account posts between two runs.

Limits

• 12 latest posts per account, the most Instagram shows without signing in. No stories, highlights, tagged posts or hashtag pages.
• The no-speech check keeps a margin: a short exclamation, distant speech or a narration over loud music can come back as no_speech, free. In our tests, a documentary trailer narrated over loud music was one of them.
• Instagram hides play counts from visitors who are not signed in: viewCount is usually empty. Likes and comments are there.
• Private accounts and age-restricted reels need a signed-in account; we do not use one, so they come back as free error records.

What it costs

$0.001 per transcript plus $0.0045 per started minute of video on Apify's free plan ($0.0034 a minute on Gold and above), and a start fee of $0.004 per run at the default memory. A 40-second reel costs $0.0055. The 6 posts of @bbcnews above cost $0.04: 5 transcripts, 7 minutes, the start fee; the reel without speech was free. Apify's free plan includes $5 of usage a month.