Instagram Reel Transcript Generator & API: Reels to Text

Instagram Reel Transcript Generator & API: Reels to Text

Turn Instagram Reels and videos into text: paste Reel links, or add whole profiles (every Reel, a date range, or only the new ones since your last run). For each Reel you get the full transcript, the hook (the words of the first 3 seconds), the language, timed segments and ready SRT and VTT subtitles, next to its caption, date, plays and likes. Speech is recognized by AI (Whisper) in 99 languages, detected for every video. Music-only Reels are free.

{ "postUrls": ["https://www.instagram.com/reel/DZ-dSUgSRiy/"], "profiles": ["garyvee"], "maxReelsPerProfile": 10 }

No Instagram login, no cookies, no sign-up on a transcript website. Use it as an Instagram transcript API: one call returns clean JSON for a list of Reels or a whole creator.

What you get

One row per Reel or video. Real example, October 2026:

ownerUsername	hook	language	wordCount	durationSeconds	billedMinutes
zachking	This is a message to all artificial intelligence.	en	50	27.9	1
garyvee	How can I keep my daughter, nine years old, away from this danger from the internet?	en	205	67.3	2
natgeo (profile)	Hey Nat Geo, nightmare scenario. You get swallowed by a sperm whale, now what?	en	432	136.2	3
luisitocomunica (profile)	Traten de seguirme.	es	98	48.6	1
natgeo	(music only: free row with the reason)		0	58.3	0

Field	Description
transcript	The spoken words as one text
hook	The words of the first 3 seconds: the opening line that keeps viewers watching
segments	[{start, end, text}] with seconds, for quotes, clips and search inside a video
subtitlesSrt, subtitlesVtt	Ready subtitles to upload to a video editor, YouTube or a player
srtFileUrl, vttFileUrl, txtFileUrl	Download links of the same as files (with Save files)
language, languageName	Detected (or chosen) spoken language: en / English
wordCount, segmentCount, speechSeconds	How much is said, and for how long
durationSeconds, billedMinutes	Video length and the minutes charged
transcriptionModel	The Whisper model that wrote the text
postUrl, shortcode, type	Link, shortcode, reel / video / carousel
caption, hashtags, postedAt	Caption, hashtags, exact posting time (UTC)
plays, likes, comments	Plays (Reels from profiles), likes and comments as Instagram shows them
audioTitle	The song or original audio of the Reel
ownerUsername, ownerFullName, ownerVerified	The account that posted it
hasTranscript, noTranscriptReason	false with the reason when a Reel has no speech, no sound or is a photo post (free)
input, inputProfile, position, scrapedAt, error	Which input the row came from; why a post was not read

How it works

1. Add links (instagram.com/reel/..., instagram.com/p/..., /tv/..., shortcodes) and / or profiles (garyvee, @garyvee, instagram.com/garyvee/, user IDs)
2. Run: each video's sound is turned into text by AI
3. Open the Transcripts view, export to CSV / Excel / JSON, or download .srt / .txt files

Input	What it does
Instagram Reel or video links	Reels, video posts, IGTV and the first video of a carousel
Instagram profiles	The Reels of each profile, newest first
Reels per profile	Latest N Reels of each profile (default 10, 0 = the whole Reels tab)
Posted after / before	Only Reels from a date range: 2026-09-01 or 7 days, 2 weeks, 1 month
Skip pinned Reels	Leave out the old Reels pinned to the top
Only new Reels since the previous run	Skip what earlier runs already transcribed: schedule it to follow creators
Spoken language	Empty = detected per video. Set en, es, hi... only if detection picks the wrong one
Max video length	Longer videos are skipped for free (default 20 minutes)
Timed segments / SRT and VTT / Save files	What goes into each row, and .srt .vtt .txt files in your storage

Why this Actor

🎯 Only speech is charged	A voice detector runs first: music-only Reels, silent videos and photo posts get a free row with the reason, and AI never invents words over a song
👤 Whole profiles	Every Reel of a creator, a date range, or only the new ones since the last run, with plays and likes next to each transcript
🪝 Hook + segments + subtitles	The opening line for content research, timestamps for quotes, SRT and VTT ready for editors
🌍 99 languages	Whisper detects the language of each video: English, Spanish, Hindi, Portuguese, Arabic and more in one run
💰 Pay per minute, not per subscription	$0.01499 per started minute of a Reel with speech; most Reels are under one minute

Follow creators automatically

1. Add the profiles you watch (competitors, clients, your own account)
2. Turn on Only new Reels since the previous run
3. Save as a task with a daily schedule: each run transcribes only the Reels posted since, ready for a Google Sheet, Slack or an AI summary through an integration

The list of transcribed Reels is kept per profile and monitor name in your account (key-value store instagram-transcript-state).

Pricing

Pay per started minute of video with speech, no subscription needed. Apify Scale and Business plans pay less:

Event	Free and Starter plans	Scale plan	Business plan
Run start	$0.001	$0.001	$0.001
Transcript, per started minute of video (a 45-second Reel = 1 minute)	$0.01499	$0.01349	$0.01199
New profile lookup (only when Instagram shows a new profile's page to a residential IP only)	$0.004	$0.004	$0.004

Task	Cost
10 Reels under a minute	~$0.15
1,000 Reels under a minute	~$14.99
Following a creator who posts 2 Reels a day, for a month	~$0.90 per month

Reels without speech, photo posts, posts that are not found and videos over the length limit are not charged.

Ready-made tasks

Open it, change the link, and run:

Task	What you get
Instagram Video to Text Converter: Reel Link to Transcript	Transcripts, hooks and SRT files of the Reels you paste

FAQ

How accurate is the transcript? It is written by OpenAI's Whisper speech recognition (large-v3-turbo; the smaller Whisper model at peak load). Clear speech comes out nearly word for word with punctuation; names, slang and heavy background music can be misheard. Check transcriptionModel in each row.

Does it read the text on screen or song lyrics? No: only spoken words. A Reel with just music gets hasTranscript: false and is free.

Which languages are supported? The 99 languages of Whisper; the language is detected for each video. If a video is mostly music with a few words, detection can pick the wrong language: set Spoken language to fix it.

Can I transcribe all Reels of an Instagram profile? Yes: add the username and set Reels per profile to 0. Private accounts cannot be read.

How are minutes counted? By the video's length, rounded up: a 45-second Reel is 1 minute, a 70-second one is 2. Only videos with speech are charged.

Do I need an Instagram account or cookies? No. The Actor reads public posts the way Instagram shows them to a visitor.

I need the video files, not the text. Use Instagram Video Downloader for MP4 files in 1080p and one ZIP.

Use with the API and AI agents

Run it via the Apify API from Python, Node.js or any HTTP client, or connect it to Claude, ChatGPT and other AI agents through the Apify MCP server: "Transcribe the last 10 Reels of @garyvee and list their hooks".

Other social media tools

Tool	What it does
Instagram Post Scraper	Any post, Reel or carousel by link with the exact posting time, likes, comments and plays
Instagram Trending Audio	Reels sounds trending right now in any country, with Reels counts and daily growth
TikTok Transcript Scraper	TikTok subtitles and AI transcripts from links or profiles, SRT and VTT
YouTube Transcript Scraper	YouTube captions of videos, channels and playlists in any language

Is it legal?

The Actor reads only public posts that Instagram shows to any visitor, without logging in. Transcripts are the creators' words: use them for research, accessibility, analysis or quotes in line with copyright law, and handle personal data such as usernames in line with GDPR and other privacy laws.