Free Video Transcript Generator
Turn Any Video Into Text
Need this at scale? Podcast Transcript API →
How to Get a Transcript From Any Video
Bring any video link
A YouTube URL, a TikTok or Reel, a podcast episode, or a direct .mp3 or .mp4 link. The source is detected for you.
The shortest path to text
If there is a caption track we read it, which is near-instant. If not, we transcribe the audio instead.
Copy, download, or automate
Take the plain text, keep timestamps for subtitles or citations, or move to the API when one at a time is not enough.
What You Get
One box for every source
Video platforms, podcasts, or a direct file URL, all through the same field.
Works when there are no captions
A missing caption track just means the audio gets transcribed instead.
Timestamped segments
Every segment carries its own start time, not one undifferentiated wall of text.
Long recordings handled
Up to four hours, chunked automatically instead of being cut short.
No sign-up, no install
Nothing to download and no account to create.
Clean output for AI pipelines
Ready for summaries, quotes, subtitles, or a prompt without cleaning it up first.
Every Source in One Box
One box, every source. A youtube.com/watch link, a TikTok or Instagram Reel, a Spotify or Apple Podcasts episode, an RSS feed, or a direct .mp3 or .mp4 URL. The source is detected from the link, so there is nothing to pick. What it cannot reach is anything behind a login or a paywall: a private upload, a members-only post, or a platform exclusive with no public file to fetch.
Need Transcripts in Your Code?
This tool is for one episode at a time. To work through a show programmatically, the podcast transcript API returns the same text as timestamped JSON from a single call, with the show and episode named in the response. Check pricing. 100 credits a month are free and only successful fetches are billed.
Why one tool for every source beats five
Getting text out of a video is a different problem on every platform, which is why most tools only solve one. YouTube usually has an auto-caption track sitting there for the taking. TikTok and Instagram mostly do not, so the audio has to be transcribed. A podcast link is not a file at all, it is a directory entry that has to be resolved back to the publisher's feed before anything can be fetched.
Those three paths produce the same thing at the end: words with timestamps. So the only real cost of them being different is that you have to know which one you need before you start. This page removes that step. Paste the link, the source is detected, and the shortest route to text is taken without you choosing.
Where it stops is access, not format. Anything public and playable can be read. Anything behind a login, a paywall, or a platform-exclusive deal has no public file to fetch, so no tool can reach it, this one included. If you can open it in a browser without signing in, it will almost certainly work here.
Frequently asked questions
What counts as a video here?
Anything with speech and a public URL. YouTube, TikTok, Instagram Reels, podcast episodes from Spotify, Apple Podcasts or an RSS feed, and direct links to media files. If you can open it in a browser without logging in, it will usually work.
Is it really free?
Yes, for one video at a time, with no account. The free tool is rate limited per hour. For bulk or programmatic use the API includes 100 free credits every month.
Can I upload a file from my computer?
Not here. This tool takes URLs, so a local file needs to be somewhere reachable first: cloud storage with a public link, or any host that serves the file directly. The API accepts direct media URLs the same way.
What if the video has no captions?
It still works. Captions are just the fastest route when they exist. Most TikToks and Reels have no caption track at all, so those are transcribed from the audio, which takes longer but produces the same shape of result.
Does the source change the accuracy?
Yes, more than the tool does. A clearly recorded podcast transcribes better than a noisy phone video with music under the speech. Multiple people talking over each other is the hardest case on any platform.
How long can a video be?
Up to four hours. Long recordings are split into chunks and processed in parallel rather than being refused for length, which is what makes full podcast episodes and conference talks practical.
Why do the timestamps not match my player exactly?
Some platforms insert ads dynamically, so the same recording is a different length for every viewer. The spoken content is identical; the offsets shift. This matters most on podcasts and least on short-form video.
Can I get subtitles out of this?
The timestamped segments are the raw material for SRT or VTT. Copy them out and format, or use the API, which returns the same segments as structured JSON so you can generate subtitle files in code.
Need this at scale? Podcast Transcript API →