Transcribe TikTok and Instagram videos to text with Python (Whisper, no API key)
Description and use
A Python script leveraging Apify and Whisper transcribes TikTok and Instagram videos in one step, returning transcripts and metadata, and can rank an entire account’s videos by views.
Problem
Short videos on TikTok and Instagram lack transcripts and captions, making it hard to extract spoken content for marketing, feedback, or trend analysis.
Outcome
The script transcribes TikTok and Instagram videos in under a minute, costs $0.006 per minute of video, and can process entire accounts, producing CSVs with dates, views, likes, and hooks.
Workflow
- Install apify-client and set APIFY_TOKEN
- Create Apify client instance
- Call the actor with run_input specifying sources (video URLs) and desired outputs
- Iterate over dataset items to retrieve transcript and metadata
- Print or store the transcript and view count
- For account ranking: call actor with profiles, maxVideosPerProfile, onlyNewVideos, and outputs
- Iterate over items, extract hook (first sentence), date, views, likes, and write to CSV
Evidence
TikTok and Instagram don't give you the transcript, and most videos have no captions.
It runs open‑source Whisper and costs $0.006 per minute of video; videos with only music are not charged.
Real output (October 2026): ... tiktok | tiktok | 335,300 views | 1 min ...
- Record date: 2026-10-01T05:08:03.578Z