<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>DEV Community: Aemon</title>
    <description>The latest articles on DEV Community by Aemon (@aemon-cz).</description>
    <link>https://dev.to/aemon-cz</link>
    <image>
      <url>https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F4244600%2F72c0386e-c92b-4ef7-bb2c-3c026996c777.png</url>
      <title>DEV Community: Aemon</title>
      <link>https://dev.to/aemon-cz</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/aemon-cz"/>
    <language>en</language>
    <item>
      <title>youtube-transcript-api RequestBlocked on AWS? 3 Ways to Get YouTube Transcripts From a Server</title>
      <dc:creator>Aemon</dc:creator>
      <pubDate>Wed, 18 Nov 2026 14:00:00 +0000</pubDate>
      <link>https://dev.to/aemon-cz/youtube-transcript-api-requestblocked-on-aws-3-ways-to-get-youtube-transcripts-from-a-server-39</link>
      <guid>https://dev.to/aemon-cz/youtube-transcript-api-requestblocked-on-aws-3-ways-to-get-youtube-transcripts-from-a-server-39</guid>
      <description>&lt;p&gt;&lt;strong&gt;TL;DR:&lt;/strong&gt; &lt;code&gt;youtube-transcript-api&lt;/code&gt; works on your laptop and then fails with &lt;code&gt;RequestBlocked&lt;/code&gt; or &lt;code&gt;IpBlocked&lt;/code&gt; as soon as it runs on AWS, Google Cloud, Azure, Heroku or a CI runner. YouTube blocks IP addresses that belong to cloud providers. You have three ways out: keep the job on a home connection, send the library's requests through a residential proxy, or call a hosted transcript API that already handles the proxies. Code for all three is below.&lt;/p&gt;

&lt;h2&gt;
  
  
  Why it works locally and fails on a server
&lt;/h2&gt;

&lt;p&gt;The library's own README says it plainly in the section &lt;a href="https://github.com/jdepoix/youtube-transcript-api#working-around-ip-bans-requestblocked-or-ipblocked-exception" rel="noopener noreferrer"&gt;Working around IP bans&lt;/a&gt;: YouTube has started blocking most IPs known to belong to cloud providers. Your home IP belongs to a consumer internet provider, so YouTube treats it like any viewer. A server IP belongs to a data center, so YouTube answers with a bot check instead of captions.&lt;/p&gt;

&lt;p&gt;I measured this while building a transcript tool. From data center IPs, YouTube's player API answered with a "sign in to confirm you're not a bot" check in 11 of 12 requests. Through residential IPs, all 12 of 12 requests returned the caption tracks. Retrying from the same server rarely helps, because the whole IP range is flagged.&lt;/p&gt;

&lt;p&gt;If you are on an older version of the library, the same block can show up as a misleading &lt;code&gt;TranscriptsDisabled&lt;/code&gt; or &lt;code&gt;NoTranscriptFound&lt;/code&gt; error for videos that clearly have captions, or as YouTube's "This request has been detected as a bot" message. Upgrade with &lt;code&gt;pip install -U youtube-transcript-api&lt;/code&gt; to get the clearer &lt;code&gt;RequestBlocked&lt;/code&gt; / &lt;code&gt;IpBlocked&lt;/code&gt; exceptions, then pick one of the fixes below.&lt;/p&gt;

&lt;h2&gt;
  
  
  Fix 1: Keep it on a home connection
&lt;/h2&gt;

&lt;p&gt;If you need transcripts for a handful of videos, or a one-off dataset, run the script on your own machine. Nothing to configure, nothing to pay.&lt;/p&gt;

&lt;p&gt;It stops working when you need it &lt;strong&gt;on a schedule&lt;/strong&gt; (a cron job, a serverless function, a CI pipeline) or &lt;strong&gt;at volume&lt;/strong&gt;. Hundreds of requests from one home IP will eventually get rate-limited too.&lt;/p&gt;

&lt;h2&gt;
  
  
  Fix 2: Route the library through a residential proxy
&lt;/h2&gt;

&lt;p&gt;&lt;code&gt;youtube-transcript-api&lt;/code&gt; accepts a proxy configuration. With any residential proxy provider that gives you an HTTP proxy URL, it looks like this:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight python"&gt;&lt;code&gt;&lt;span class="k"&gt;from&lt;/span&gt; &lt;span class="n"&gt;youtube_transcript_api&lt;/span&gt; &lt;span class="k"&gt;import&lt;/span&gt; &lt;span class="nc"&gt;YouTubeTranscriptApi&lt;/span&gt;
&lt;span class="k"&gt;from&lt;/span&gt; &lt;span class="n"&gt;youtube_transcript_api&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="n"&gt;proxies&lt;/span&gt; &lt;span class="k"&gt;import&lt;/span&gt; &lt;span class="nc"&gt;GenericProxyConfig&lt;/span&gt;

&lt;span class="n"&gt;proxy&lt;/span&gt; &lt;span class="o"&gt;=&lt;/span&gt; &lt;span class="s2"&gt;"http://USERNAME:PASSWORD@your-residential-proxy-host:PORT"&lt;/span&gt;
&lt;span class="n"&gt;api&lt;/span&gt; &lt;span class="o"&gt;=&lt;/span&gt; &lt;span class="nc"&gt;YouTubeTranscriptApi&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="n"&gt;proxy_config&lt;/span&gt;&lt;span class="o"&gt;=&lt;/span&gt;&lt;span class="nc"&gt;GenericProxyConfig&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="n"&gt;http_url&lt;/span&gt;&lt;span class="o"&gt;=&lt;/span&gt;&lt;span class="n"&gt;proxy&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="n"&gt;https_url&lt;/span&gt;&lt;span class="o"&gt;=&lt;/span&gt;&lt;span class="n"&gt;proxy&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt;

&lt;span class="n"&gt;transcript&lt;/span&gt; &lt;span class="o"&gt;=&lt;/span&gt; &lt;span class="n"&gt;api&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="nf"&gt;fetch&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="s2"&gt;"aircAruvnKk"&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt;
&lt;span class="k"&gt;print&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="n"&gt;transcript&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="nf"&gt;to_raw_data&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt;&lt;span class="p"&gt;[&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt;&lt;span class="mi"&gt;3&lt;/span&gt;&lt;span class="p"&gt;]&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;The library also ships a ready-made config for one specific provider; the README section linked above shows it.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;Good:&lt;/strong&gt; you keep the free, open-source library and full control.&lt;br&gt;
&lt;strong&gt;Watch out for:&lt;/strong&gt; you need a proxy subscription billed per GB, the proxy must be &lt;em&gt;residential&lt;/em&gt; (data center proxies get the same bot check), and you still write your own retries for the requests that land on a flagged IP. Channels, playlists and search are not part of the library, so listing a channel's videos is another piece you build yourself.&lt;/p&gt;
&lt;h2&gt;
  
  
  Fix 3: Call a hosted transcript API
&lt;/h2&gt;

&lt;p&gt;If you would rather not run proxies at all, use a service that does it for you. I built one for exactly this: &lt;a href="https://apify.com/aemon/youtube-transcript-scraper" rel="noopener noreferrer"&gt;YouTube Transcript Scraper&lt;/a&gt;, an Actor on the Apify platform. You send video links, and you get back JSON with the transcript. The residential proxies and the retries happen on its side.&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight python"&gt;&lt;code&gt;&lt;span class="k"&gt;from&lt;/span&gt; &lt;span class="n"&gt;apify_client&lt;/span&gt; &lt;span class="k"&gt;import&lt;/span&gt; &lt;span class="nc"&gt;ApifyClient&lt;/span&gt;

&lt;span class="n"&gt;client&lt;/span&gt; &lt;span class="o"&gt;=&lt;/span&gt; &lt;span class="nc"&gt;ApifyClient&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="s2"&gt;"YOUR_APIFY_TOKEN"&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt;
&lt;span class="n"&gt;run&lt;/span&gt; &lt;span class="o"&gt;=&lt;/span&gt; &lt;span class="n"&gt;client&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="nf"&gt;actor&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="s2"&gt;"aemon/youtube-transcript-scraper"&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="nf"&gt;call&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="n"&gt;run_input&lt;/span&gt;&lt;span class="o"&gt;=&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;
    &lt;span class="s2"&gt;"urls"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="p"&gt;[&lt;/span&gt;&lt;span class="s2"&gt;"https://www.youtube.com/watch?v=aircAruvnKk"&lt;/span&gt;&lt;span class="p"&gt;]&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
&lt;span class="p"&gt;}&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt;
&lt;span class="k"&gt;for&lt;/span&gt; &lt;span class="n"&gt;item&lt;/span&gt; &lt;span class="k"&gt;in&lt;/span&gt; &lt;span class="n"&gt;client&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="nf"&gt;dataset&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="n"&gt;run&lt;/span&gt;&lt;span class="p"&gt;[&lt;/span&gt;&lt;span class="s2"&gt;"defaultDatasetId"&lt;/span&gt;&lt;span class="p"&gt;]&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="nf"&gt;iterate_items&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt;
    &lt;span class="k"&gt;print&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="n"&gt;item&lt;/span&gt;&lt;span class="p"&gt;[&lt;/span&gt;&lt;span class="s2"&gt;"title"&lt;/span&gt;&lt;span class="p"&gt;]&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="n"&gt;item&lt;/span&gt;&lt;span class="p"&gt;[&lt;/span&gt;&lt;span class="s2"&gt;"wordCount"&lt;/span&gt;&lt;span class="p"&gt;]&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt;
    &lt;span class="k"&gt;print&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="n"&gt;item&lt;/span&gt;&lt;span class="p"&gt;[&lt;/span&gt;&lt;span class="s2"&gt;"text"&lt;/span&gt;&lt;span class="p"&gt;]&lt;/span&gt;&lt;span class="p"&gt;[&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt;&lt;span class="mi"&gt;200&lt;/span&gt;&lt;span class="p"&gt;]&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;This is the real result for that video (shortened):&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight json"&gt;&lt;code&gt;&lt;span class="p"&gt;{&lt;/span&gt;
    &lt;span class="nl"&gt;"status"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="s2"&gt;"ok"&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
    &lt;span class="nl"&gt;"videoId"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="s2"&gt;"aircAruvnKk"&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
    &lt;span class="nl"&gt;"title"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="s2"&gt;"But what is a neural network? | Deep learning chapter 1"&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
    &lt;span class="nl"&gt;"channel"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="s2"&gt;"3Blue1Brown"&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
    &lt;span class="nl"&gt;"durationSec"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="mi"&gt;1120&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
    &lt;span class="nl"&gt;"views"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="mi"&gt;25266486&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
    &lt;span class="nl"&gt;"language"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="s2"&gt;"en"&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
    &lt;span class="nl"&gt;"captionType"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="s2"&gt;"manual"&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
    &lt;span class="nl"&gt;"wordCount"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="mi"&gt;3357&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
    &lt;span class="nl"&gt;"text"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="s2"&gt;"This is a 3. It's sloppily written and rendered at an extremely low resolution of 28x28 pixels, but your brain has no trouble recognizing it as a 3..."&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
    &lt;span class="nl"&gt;"segments"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="p"&gt;[&lt;/span&gt;
        &lt;span class="p"&gt;{&lt;/span&gt; &lt;span class="nl"&gt;"start"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="mf"&gt;4.22&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="nl"&gt;"end"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="mf"&gt;5.4&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="nl"&gt;"text"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="s2"&gt;"This is a 3."&lt;/span&gt; &lt;span class="p"&gt;}&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
        &lt;span class="p"&gt;{&lt;/span&gt; &lt;span class="nl"&gt;"start"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="mf"&gt;6.06&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="nl"&gt;"end"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="mf"&gt;10.71&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="nl"&gt;"text"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="s2"&gt;"It's sloppily written and rendered at an extremely low resolution of 28x28 pixels,"&lt;/span&gt; &lt;span class="p"&gt;}&lt;/span&gt;
    &lt;span class="p"&gt;]&lt;/span&gt;
&lt;span class="p"&gt;}&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;The run took under 10 seconds and moved about 51 KB through the proxy for an 18-minute video, because it reads only the caption data and never downloads the video.&lt;/p&gt;

&lt;h3&gt;
  
  
  A drop-in replacement for &lt;code&gt;fetch()&lt;/code&gt;
&lt;/h3&gt;

&lt;p&gt;If your code already expects the library's format (&lt;code&gt;text&lt;/code&gt;, &lt;code&gt;start&lt;/code&gt;, &lt;code&gt;duration&lt;/code&gt;), wrap the call in a small function and keep the rest of your code as it is:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight python"&gt;&lt;code&gt;&lt;span class="k"&gt;from&lt;/span&gt; &lt;span class="n"&gt;apify_client&lt;/span&gt; &lt;span class="k"&gt;import&lt;/span&gt; &lt;span class="nc"&gt;ApifyClient&lt;/span&gt;

&lt;span class="n"&gt;client&lt;/span&gt; &lt;span class="o"&gt;=&lt;/span&gt; &lt;span class="nc"&gt;ApifyClient&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="s2"&gt;"YOUR_APIFY_TOKEN"&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt;

&lt;span class="k"&gt;def&lt;/span&gt; &lt;span class="nf"&gt;fetch_transcript&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="n"&gt;video_id&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="kt"&gt;str&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt; &lt;span class="o"&gt;-&amp;gt;&lt;/span&gt; &lt;span class="kt"&gt;list&lt;/span&gt;&lt;span class="p"&gt;[&lt;/span&gt;&lt;span class="kt"&gt;dict&lt;/span&gt;&lt;span class="p"&gt;]&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt;
    &lt;span class="n"&gt;run&lt;/span&gt; &lt;span class="o"&gt;=&lt;/span&gt; &lt;span class="n"&gt;client&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="nf"&gt;actor&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="s2"&gt;"aemon/youtube-transcript-scraper"&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="nf"&gt;call&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;
        &lt;span class="n"&gt;run_input&lt;/span&gt;&lt;span class="o"&gt;=&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="s2"&gt;"urls"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="p"&gt;[&lt;/span&gt;&lt;span class="n"&gt;video_id&lt;/span&gt;&lt;span class="p"&gt;]&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="s2"&gt;"includeSegments"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="kc"&gt;True&lt;/span&gt;&lt;span class="p"&gt;}&lt;/span&gt;
    &lt;span class="p"&gt;)&lt;/span&gt;
    &lt;span class="n"&gt;item&lt;/span&gt; &lt;span class="o"&gt;=&lt;/span&gt; &lt;span class="nb"&gt;next&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="n"&gt;client&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="nf"&gt;dataset&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="n"&gt;run&lt;/span&gt;&lt;span class="p"&gt;[&lt;/span&gt;&lt;span class="s2"&gt;"defaultDatasetId"&lt;/span&gt;&lt;span class="p"&gt;]&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="nf"&gt;iterate_items&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt;
    &lt;span class="k"&gt;if&lt;/span&gt; &lt;span class="n"&gt;item&lt;/span&gt;&lt;span class="p"&gt;[&lt;/span&gt;&lt;span class="s2"&gt;"status"&lt;/span&gt;&lt;span class="p"&gt;]&lt;/span&gt; &lt;span class="o"&gt;!=&lt;/span&gt; &lt;span class="s2"&gt;"ok"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt;
        &lt;span class="k"&gt;raise&lt;/span&gt; &lt;span class="nc"&gt;RuntimeError&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="n"&gt;f&lt;/span&gt;&lt;span class="s1"&gt;'{item["errorCode"]}: {item["error"]}'&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt;
    &lt;span class="k"&gt;return&lt;/span&gt; &lt;span class="p"&gt;[&lt;/span&gt;
        &lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="s2"&gt;"text"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="n"&gt;s&lt;/span&gt;&lt;span class="p"&gt;[&lt;/span&gt;&lt;span class="s2"&gt;"text"&lt;/span&gt;&lt;span class="p"&gt;]&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="s2"&gt;"start"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="n"&gt;s&lt;/span&gt;&lt;span class="p"&gt;[&lt;/span&gt;&lt;span class="s2"&gt;"start"&lt;/span&gt;&lt;span class="p"&gt;]&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="s2"&gt;"duration"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="nb"&gt;round&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="n"&gt;s&lt;/span&gt;&lt;span class="p"&gt;[&lt;/span&gt;&lt;span class="s2"&gt;"end"&lt;/span&gt;&lt;span class="p"&gt;]&lt;/span&gt; &lt;span class="o"&gt;-&lt;/span&gt; &lt;span class="n"&gt;s&lt;/span&gt;&lt;span class="p"&gt;[&lt;/span&gt;&lt;span class="s2"&gt;"start"&lt;/span&gt;&lt;span class="p"&gt;]&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="mi"&gt;2&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt;&lt;span class="p"&gt;}&lt;/span&gt;
        &lt;span class="k"&gt;for&lt;/span&gt; &lt;span class="n"&gt;s&lt;/span&gt; &lt;span class="k"&gt;in&lt;/span&gt; &lt;span class="n"&gt;item&lt;/span&gt;&lt;span class="p"&gt;[&lt;/span&gt;&lt;span class="s2"&gt;"segments"&lt;/span&gt;&lt;span class="p"&gt;]&lt;/span&gt;
    &lt;span class="p"&gt;]&lt;/span&gt;

&lt;span class="k"&gt;print&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="nf"&gt;fetch_transcript&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="s2"&gt;"aircAruvnKk"&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt;&lt;span class="p"&gt;[&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt;&lt;span class="mi"&gt;3&lt;/span&gt;&lt;span class="p"&gt;]&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Bare 11-character video IDs work as input, as do watch, Shorts, live, embed and &lt;code&gt;youtu.be&lt;/code&gt; links.&lt;/p&gt;

&lt;h3&gt;
  
  
  Things the library can't do
&lt;/h3&gt;

&lt;p&gt;Because it is a service rather than a library, a few extras come with it:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Whole channels and playlists&lt;/strong&gt;: pass &lt;code&gt;https://www.youtube.com/@3blue1brown&lt;/code&gt; and set how many of the newest videos you want.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;New videos only&lt;/strong&gt;: &lt;code&gt;publishedAfter: "2026-11-01"&lt;/code&gt; returns just what a channel published since that date, which makes a weekly cron job trivial.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;YouTube search&lt;/strong&gt;: &lt;code&gt;searchQueries: ["rag tutorial"]&lt;/code&gt; transcribes the top results for a keyword.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;AI-dubbed videos&lt;/strong&gt;: when a creator uses YouTube's AI dubbing, the transcript comes from the original language, not a dub.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Optional AI summary&lt;/strong&gt; with chapters, key points and sponsors, and &lt;strong&gt;SRT/WebVTT&lt;/strong&gt; subtitles.&lt;/li&gt;
&lt;/ul&gt;

&lt;h3&gt;
  
  
  What it costs
&lt;/h3&gt;

&lt;p&gt;$0.002 per video ($2 per 1,000), with the proxy included. Videos without captions, private and removed videos are not charged; they come back as a row with an &lt;code&gt;errorCode&lt;/code&gt; such as &lt;code&gt;NO_CAPTIONS&lt;/code&gt;. Apify's free plan includes monthly credit, so you can test it on your own videos before paying anything.&lt;/p&gt;

&lt;h2&gt;
  
  
  Which fix should you pick?
&lt;/h2&gt;

&lt;div class="table-wrapper-paragraph"&gt;&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;&lt;/th&gt;
&lt;th&gt;Home connection&lt;/th&gt;
&lt;th&gt;Library + residential proxy&lt;/th&gt;
&lt;th&gt;Hosted API&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;Works on AWS, GCP, Heroku, CI&lt;/td&gt;
&lt;td&gt;❌&lt;/td&gt;
&lt;td&gt;✅&lt;/td&gt;
&lt;td&gt;✅&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Setup&lt;/td&gt;
&lt;td&gt;None&lt;/td&gt;
&lt;td&gt;Proxy account, retries&lt;/td&gt;
&lt;td&gt;API token&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Cost&lt;/td&gt;
&lt;td&gt;Free&lt;/td&gt;
&lt;td&gt;Proxy plan, billed per GB&lt;/td&gt;
&lt;td&gt;$0.002 per video&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Channels, playlists, search&lt;/td&gt;
&lt;td&gt;❌&lt;/td&gt;
&lt;td&gt;Build it yourself&lt;/td&gt;
&lt;td&gt;✅ Built in&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Best for&lt;/td&gt;
&lt;td&gt;A few videos, one-off&lt;/td&gt;
&lt;td&gt;Full control, high volume&lt;/td&gt;
&lt;td&gt;Pipelines, agents, cron jobs&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;&lt;/div&gt;

&lt;p&gt;If you transcribe a few videos by hand, stay local. If you run millions of videos and want to own every part, a proxy plan with the library is the cheapest at that scale. For everything in between, especially scheduled jobs and AI agents, a hosted API saves you the proxy work.&lt;/p&gt;

&lt;h2&gt;
  
  
  Bonus: YouTube transcripts inside an AI agent
&lt;/h2&gt;

&lt;p&gt;The Actor also works as a tool for Claude, ChatGPT, Cursor and other MCP clients through the &lt;a href="https://docs.apify.com/platform/integrations/mcp" rel="noopener noreferrer"&gt;Apify MCP server&lt;/a&gt;. Add &lt;code&gt;https://mcp.apify.com?tools=aemon/youtube-transcript-scraper&lt;/code&gt; to your MCP config and ask &lt;em&gt;"Get the transcript of this video and summarise it in five bullet points"&lt;/em&gt;. Agents can even pay per run on their own through Apify's &lt;a href="https://docs.apify.com/agents/payments" rel="noopener noreferrer"&gt;agentic payments&lt;/a&gt;.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;Which of the three are you using today?&lt;/strong&gt; Tell me in the comments, especially if you have found another way around the block.&lt;/p&gt;

&lt;p&gt;&lt;em&gt;Disclosure: I'm the author of YouTube Transcript Scraper, published on Apify under my agency &lt;a href="https://www.aemon.cz/" rel="noopener noreferrer"&gt;Aemon&lt;/a&gt;. The measurements above come from my own test runs in November 2026.&lt;/em&gt;&lt;/p&gt;

</description>
      <category>python</category>
      <category>youtube</category>
      <category>ai</category>
      <category>tutorial</category>
    </item>
    <item>
      <title>How to Transcribe Instagram Reels and TikTok Videos for AI (in Bulk)</title>
      <dc:creator>Aemon</dc:creator>
      <pubDate>Tue, 17 Nov 2026 14:00:00 +0000</pubDate>
      <link>https://dev.to/aemon-cz/how-to-transcribe-instagram-reels-and-tiktok-videos-for-ai-in-bulk-1g52</link>
      <guid>https://dev.to/aemon-cz/how-to-transcribe-instagram-reels-and-tiktok-videos-for-ai-in-bulk-1g52</guid>
      <description>&lt;p&gt;&lt;strong&gt;TL;DR:&lt;/strong&gt; LLMs can't watch videos. To use Reels and TikToks in ChatGPT, Claude or a RAG pipeline, you need them as text. The reliable way: use platform captions when they exist, transcribe the audio with Whisper when they don't, and return structured JSON. Below you'll find a no-code option, JavaScript and Python examples, and a pipeline for transcribing a whole niche.&lt;/p&gt;

&lt;p&gt;Short-form video is where a lot of today's knowledge, opinions and marketing lives: product demos on TikTok, expert tips in Instagram Reels, announcements on X. But none of it is searchable or usable by AI until it's text.&lt;/p&gt;

&lt;p&gt;This guide shows how to get clean, timestamped transcripts from &lt;strong&gt;Instagram, TikTok, X, Facebook and YouTube&lt;/strong&gt; videos, one at a time or thousands at once, without running any servers yourself.&lt;/p&gt;

&lt;h2&gt;
  
  
  Why transcribing social videos is harder than it looks
&lt;/h2&gt;

&lt;p&gt;If you've tried it, you've probably hit some of these:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Every platform is different.&lt;/strong&gt; YouTube often has captions, TikTok sometimes does, Instagram and X almost never do.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Downloading is the hard part.&lt;/strong&gt; Social networks block scripts, links expire, and formats change every few weeks.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Speech-to-text has to be good.&lt;/strong&gt; Reels are full of music, accents and fast talking.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;It needs to scale.&lt;/strong&gt; Researching a niche means hundreds of videos, not one.&lt;/li&gt;
&lt;/ul&gt;

&lt;h2&gt;
  
  
  The approach: captions first, Whisper second
&lt;/h2&gt;

&lt;p&gt;The most reliable pipeline looks like this:&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;
&lt;strong&gt;Resolve the video URL&lt;/strong&gt; and read its metadata (author, caption, views, publish date).&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Use the platform's own captions&lt;/strong&gt; if they exist in the spoken language. They're free and instant.&lt;/li&gt;
&lt;li&gt;Otherwise &lt;strong&gt;download only the audio track&lt;/strong&gt; (a few hundred kilobytes for a Reel instead of the whole video) and transcribe it with &lt;strong&gt;Whisper large-v3-turbo&lt;/strong&gt;, which handles music, accents and 90+ languages well.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Return structured JSON&lt;/strong&gt; with the full text, timestamped segments and optional SRT/VTT subtitles.&lt;/li&gt;
&lt;li&gt;Optionally, &lt;strong&gt;read the text shown on screen&lt;/strong&gt; (many Reels carry their message in captions over music) and &lt;strong&gt;let an LLM break the video down&lt;/strong&gt;: summary, hook, key points and call to action.&lt;/li&gt;
&lt;/ol&gt;

&lt;p&gt;You can build this yourself with &lt;code&gt;yt-dlp&lt;/code&gt;, &lt;code&gt;ffmpeg&lt;/code&gt; and a Whisper API. If you'd rather skip the maintenance (the platforms change constantly), the &lt;strong&gt;Video to Text&lt;/strong&gt; Actor on Apify does exactly this, and I'll use it in the examples below:&lt;/p&gt;


&lt;div class="crayons-card c-embed text-styles text-styles--secondary"&gt;
    &lt;div class="c-embed__content"&gt;
        &lt;div class="c-embed__cover"&gt;
          &lt;a href="https://apify.com/aemon/video-to-text-for-ai" class="c-link align-middle" rel="noopener noreferrer"&gt;
            &lt;img alt="" src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto/https%3A%2F%2Fapify.com%2Fog-image%2Factor%3FactorName%3DSocial%2BMedia%2BVideo%2Bto%2BText%253A%2BTikTok%252C%2BInstagram%252C%2BX%2BTranscripts%26uniqueName%3Daemon%252Fvideo-to-text-for-ai%26categories%3DAI%252CVIDEOS%252CSOCIAL_MEDIA%26users%3D2%26runs%3D77%26pictureUrl%3Dhttps%253A%252F%252Fapify-image-uploads-prod.s3.us-east-1.amazonaws.com%252FQiLvOjZNXpKDelkSa-actor-TsMw27QGqhpvAmWTL-h606ziyhvN-icon.png%26authorName%3DAemon%26userPictureUrl%3Dhttps%253A%252F%252Fimages.apifyusercontent.com%252FTOde8UwT8tRAse33Sd-Z6W0evLcJXshxhRS9gUfKeco%252Frs%253Afill%253A224%253A224%252Fcb%253A1%252FaHR0cHM6Ly9hcGlmeS1pbWFnZS11cGxvYWRzLXByb2QuczMudXMtZWFzdC0xLmFtYXpvbmF3cy5jb20vUWlMdk9qWk5YcEtEZWxrU2EtcHJvZmlsZS1TS1FIbWtmbW9GLUFlbW9uLnBuZw.webp" height="420" class="m-0" width="800"&gt;
          &lt;/a&gt;
        &lt;/div&gt;
      &lt;div class="c-embed__body"&gt;
        &lt;h2 class="fs-xl lh-tight"&gt;
          &lt;a href="https://apify.com/aemon/video-to-text-for-ai" rel="noopener noreferrer" class="c-link"&gt;
            Instagram, TikTok &amp;amp; X Transcript API for AI · Apify
          &lt;/a&gt;
        &lt;/h2&gt;
          &lt;p class="truncate-at-3"&gt;
            Transcribe Instagram, TikTok, X and Facebook videos to text with timestamps, on-screen text and AI summary, hook and CTA. AI-ready JSON, pay per success.
          &lt;/p&gt;
        &lt;div class="color-secondary fs-s flex items-center"&gt;
            &lt;img alt="favicon" class="c-embed__favicon m-0 mr-2 radius-0" src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fapify.com%2Ffavicon.ico%3Ffavicon.2653ilt32bing.ico%3Fdpl%3D7faf285de8" width="48" height="48"&gt;
          apify.com
        &lt;/div&gt;
      &lt;/div&gt;
    &lt;/div&gt;
&lt;/div&gt;


&lt;h2&gt;
  
  
  Option 1: No code (2 minutes)
&lt;/h2&gt;

&lt;ol&gt;
&lt;li&gt;Open &lt;a href="https://apify.com/aemon/video-to-text-for-ai" rel="noopener noreferrer"&gt;Video to Text: Instagram, TikTok &amp;amp; X Transcripts&lt;/a&gt; and click &lt;strong&gt;Try for free&lt;/strong&gt;.&lt;/li&gt;
&lt;li&gt;Paste video links, one per line. You can mix platforms. These two are real, so you can try them right away:
&lt;/li&gt;
&lt;/ol&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;https://www.instagram.com/reel/DMflZnXOKUj
https://www.tiktok.com/@apifylife/video/7338085038258457889
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;ol&gt;
&lt;li&gt;Optional: turn on &lt;strong&gt;AI analysis&lt;/strong&gt; to also get a summary, the hook, key points and the call to action of every video, or pick a language in &lt;strong&gt;Translate transcript to&lt;/strong&gt;.&lt;/li&gt;
&lt;li&gt;Click &lt;strong&gt;Start&lt;/strong&gt;. A few seconds later you get one row per video with the transcript, author, views, likes and publish date.&lt;/li&gt;
&lt;li&gt;Export it as CSV, JSON or Excel, or send it straight to Google Sheets.&lt;/li&gt;
&lt;/ol&gt;

&lt;p&gt;Only need one platform? The single-platform editions run the same engine for &lt;strong&gt;$5 per 1,000 videos&lt;/strong&gt; instead of $6: &lt;a href="https://apify.com/aemon/instagram-reels-transcript-scraper" rel="noopener noreferrer"&gt;Instagram Reels Transcript&lt;/a&gt; and &lt;a href="https://apify.com/aemon/tiktok-transcript-scraper" rel="noopener noreferrer"&gt;TikTok Transcript&lt;/a&gt;, which also takes whole TikTok profiles.&lt;/p&gt;

&lt;p&gt;For the NASA Hubble Reel above, the transcript starts like this:&lt;/p&gt;

&lt;blockquote&gt;
&lt;p&gt;"Just a few years after its launch in 1990, NASA's Hubble Space Telescope made a groundbreaking discovery..."&lt;/p&gt;
&lt;/blockquote&gt;

&lt;h2&gt;
  
  
  Option 2: JavaScript
&lt;/h2&gt;



&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight javascript"&gt;&lt;code&gt;&lt;span class="k"&gt;import&lt;/span&gt; &lt;span class="p"&gt;{&lt;/span&gt; &lt;span class="nc"&gt;ApifyClient&lt;/span&gt; &lt;span class="p"&gt;}&lt;/span&gt; &lt;span class="k"&gt;from&lt;/span&gt; &lt;span class="s1"&gt;'apify-client'&lt;/span&gt;&lt;span class="p"&gt;;&lt;/span&gt;

&lt;span class="kd"&gt;const&lt;/span&gt; &lt;span class="n"&gt;client&lt;/span&gt; &lt;span class="o"&gt;=&lt;/span&gt; &lt;span class="k"&gt;new&lt;/span&gt; &lt;span class="nc"&gt;ApifyClient&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt; &lt;span class="n"&gt;token&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="nb"&gt;process&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="n"&gt;env&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="no"&gt;APIFY_TOKEN&lt;/span&gt; &lt;span class="p"&gt;}&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt;&lt;span class="p"&gt;;&lt;/span&gt;

&lt;span class="kd"&gt;const&lt;/span&gt; &lt;span class="n"&gt;run&lt;/span&gt; &lt;span class="o"&gt;=&lt;/span&gt; &lt;span class="k"&gt;await&lt;/span&gt; &lt;span class="n"&gt;client&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="nf"&gt;actor&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="s1"&gt;'aemon/video-to-text-for-ai'&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="nf"&gt;call&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;
    &lt;span class="n"&gt;urls&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="p"&gt;[&lt;/span&gt;
        &lt;span class="s1"&gt;'https://www.instagram.com/reel/DMflZnXOKUj'&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
        &lt;span class="s1"&gt;'https://www.tiktok.com/@apifylife/video/7338085038258457889'&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
    &lt;span class="p"&gt;]&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
    &lt;span class="n"&gt;subtitleFormat&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="s1"&gt;'srt'&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
    &lt;span class="n"&gt;aiAnalysis&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="kc"&gt;true&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="c1"&gt;// summary, hook, key points and call to action&lt;/span&gt;
&lt;span class="p"&gt;}&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt;&lt;span class="p"&gt;;&lt;/span&gt;

&lt;span class="kd"&gt;const&lt;/span&gt; &lt;span class="p"&gt;{&lt;/span&gt; &lt;span class="n"&gt;items&lt;/span&gt; &lt;span class="p"&gt;}&lt;/span&gt; &lt;span class="o"&gt;=&lt;/span&gt; &lt;span class="k"&gt;await&lt;/span&gt; &lt;span class="n"&gt;client&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="nf"&gt;dataset&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="n"&gt;run&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="n"&gt;defaultDatasetId&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="nf"&gt;listItems&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt;&lt;span class="p"&gt;;&lt;/span&gt;
&lt;span class="k"&gt;for&lt;/span&gt; &lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="kd"&gt;const&lt;/span&gt; &lt;span class="n"&gt;video&lt;/span&gt; &lt;span class="k"&gt;of&lt;/span&gt; &lt;span class="n"&gt;items&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt; &lt;span class="p"&gt;{&lt;/span&gt;
    &lt;span class="k"&gt;if&lt;/span&gt; &lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="n"&gt;video&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="n"&gt;status&lt;/span&gt; &lt;span class="o"&gt;===&lt;/span&gt; &lt;span class="s1"&gt;'ok'&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt; &lt;span class="nb"&gt;console&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="nf"&gt;log&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="n"&gt;video&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="n"&gt;platform&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="n"&gt;video&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="n"&gt;author&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="n"&gt;video&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="n"&gt;aiAnalysis&lt;/span&gt;&lt;span class="o"&gt;?&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="n"&gt;hook&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="n"&gt;video&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="n"&gt;text&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt;&lt;span class="p"&gt;;&lt;/span&gt;
    &lt;span class="k"&gt;else&lt;/span&gt; &lt;span class="nb"&gt;console&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="nf"&gt;log&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="s1"&gt;'Failed:'&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="n"&gt;video&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="n"&gt;url&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="n"&gt;video&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="n"&gt;errorCode&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt;&lt;span class="p"&gt;;&lt;/span&gt;
&lt;span class="p"&gt;}&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;h2&gt;
  
  
  Option 3: Python
&lt;/h2&gt;



&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight python"&gt;&lt;code&gt;&lt;span class="k"&gt;from&lt;/span&gt; &lt;span class="n"&gt;apify_client&lt;/span&gt; &lt;span class="k"&gt;import&lt;/span&gt; &lt;span class="nc"&gt;ApifyClient&lt;/span&gt;

&lt;span class="n"&gt;client&lt;/span&gt; &lt;span class="o"&gt;=&lt;/span&gt; &lt;span class="nc"&gt;ApifyClient&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="s2"&gt;"YOUR_APIFY_TOKEN"&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt;
&lt;span class="n"&gt;run&lt;/span&gt; &lt;span class="o"&gt;=&lt;/span&gt; &lt;span class="n"&gt;client&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="nf"&gt;actor&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="s2"&gt;"aemon/video-to-text-for-ai"&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="nf"&gt;call&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;
    &lt;span class="n"&gt;run_input&lt;/span&gt;&lt;span class="o"&gt;=&lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="s2"&gt;"urls"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="p"&gt;[&lt;/span&gt;&lt;span class="s2"&gt;"https://www.tiktok.com/@apifylife/video/7338085038258457889"&lt;/span&gt;&lt;span class="p"&gt;]&lt;/span&gt;&lt;span class="p"&gt;}&lt;/span&gt;
&lt;span class="p"&gt;)&lt;/span&gt;
&lt;span class="k"&gt;for&lt;/span&gt; &lt;span class="n"&gt;video&lt;/span&gt; &lt;span class="k"&gt;in&lt;/span&gt; &lt;span class="n"&gt;client&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="nf"&gt;dataset&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="n"&gt;run&lt;/span&gt;&lt;span class="p"&gt;[&lt;/span&gt;&lt;span class="s2"&gt;"defaultDatasetId"&lt;/span&gt;&lt;span class="p"&gt;]&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="nf"&gt;iterate_items&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt;
    &lt;span class="k"&gt;print&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="n"&gt;video&lt;/span&gt;&lt;span class="p"&gt;[&lt;/span&gt;&lt;span class="s2"&gt;"author"&lt;/span&gt;&lt;span class="p"&gt;]&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="n"&gt;video&lt;/span&gt;&lt;span class="p"&gt;[&lt;/span&gt;&lt;span class="s2"&gt;"wordCount"&lt;/span&gt;&lt;span class="p"&gt;]&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="n"&gt;video&lt;/span&gt;&lt;span class="p"&gt;[&lt;/span&gt;&lt;span class="s2"&gt;"text"&lt;/span&gt;&lt;span class="p"&gt;]&lt;/span&gt;&lt;span class="p"&gt;[&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt;&lt;span class="mi"&gt;200&lt;/span&gt;&lt;span class="p"&gt;]&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;h2&gt;
  
  
  What the output looks like
&lt;/h2&gt;



&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight json"&gt;&lt;code&gt;&lt;span class="p"&gt;{&lt;/span&gt;
    &lt;span class="nl"&gt;"status"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="s2"&gt;"ok"&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
    &lt;span class="nl"&gt;"platform"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="s2"&gt;"instagram"&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
    &lt;span class="nl"&gt;"author"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="s2"&gt;"Hubble Space Telescope"&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
    &lt;span class="nl"&gt;"publishedAt"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="s2"&gt;"2025-07-24T14:16:39.000Z"&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
    &lt;span class="nl"&gt;"durationSec"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="mi"&gt;152&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
    &lt;span class="nl"&gt;"likes"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="mi"&gt;11256&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
    &lt;span class="nl"&gt;"language"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="s2"&gt;"en"&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
    &lt;span class="nl"&gt;"transcriptSource"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="s2"&gt;"whisper"&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
    &lt;span class="nl"&gt;"text"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="s2"&gt;"Just a few years after its launch in 1990, NASA's Hubble Space Telescope made a groundbreaking discovery..."&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
    &lt;span class="nl"&gt;"wordCount"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="mi"&gt;307&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
    &lt;span class="nl"&gt;"segments"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="p"&gt;[&lt;/span&gt;
        &lt;span class="p"&gt;{&lt;/span&gt; &lt;span class="nl"&gt;"start"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="mi"&gt;0&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="nl"&gt;"end"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="mf"&gt;7.76&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="nl"&gt;"text"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="s2"&gt;"Just a few years after its launch in 1990, NASA's Hubble Space Telescope made a groundbreaking discovery."&lt;/span&gt; &lt;span class="p"&gt;}&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
        &lt;span class="p"&gt;{&lt;/span&gt; &lt;span class="nl"&gt;"start"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="mf"&gt;8.52&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="nl"&gt;"end"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="mf"&gt;14.34&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="nl"&gt;"text"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="s2"&gt;"Nearly every galaxy in the universe appears to have a supermassive black hole at its center."&lt;/span&gt; &lt;span class="p"&gt;}&lt;/span&gt;
    &lt;span class="p"&gt;]&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
    &lt;span class="nl"&gt;"aiAnalysis"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="p"&gt;{&lt;/span&gt;
        &lt;span class="nl"&gt;"summary"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="s2"&gt;"The video explains why intermediate-mass black holes are hard to find and how Hubble and Chandra caught one tearing apart a star in galaxy NGC 6099."&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
        &lt;span class="nl"&gt;"hook"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="s2"&gt;"Just a few years after its launch in 1990, NASA's Hubble Space Telescope made a groundbreaking discovery."&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
        &lt;span class="nl"&gt;"keyPoints"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="p"&gt;[&lt;/span&gt;&lt;span class="s2"&gt;"Nearly every galaxy has a supermassive black hole"&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="s2"&gt;"Intermediate-mass black holes only shine when they consume a star"&lt;/span&gt;&lt;span class="p"&gt;]&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
        &lt;span class="nl"&gt;"callToAction"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="kc"&gt;null&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
        &lt;span class="nl"&gt;"topics"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="p"&gt;[&lt;/span&gt;&lt;span class="s2"&gt;"black holes"&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="s2"&gt;"hubble"&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="s2"&gt;"astronomy"&lt;/span&gt;&lt;span class="p"&gt;]&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
        &lt;span class="nl"&gt;"contentType"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="s2"&gt;"educational"&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
        &lt;span class="nl"&gt;"sentiment"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="s2"&gt;"positive"&lt;/span&gt;
    &lt;span class="p"&gt;}&lt;/span&gt;
&lt;span class="p"&gt;}&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Videos that are private, removed or have no audio come back as rows with &lt;code&gt;"status": "error"&lt;/code&gt; and an error code, so a batch never silently loses links.&lt;/p&gt;

&lt;h2&gt;
  
  
  Doing it at scale: from a creator or hashtag to transcripts
&lt;/h2&gt;

&lt;p&gt;For a single creator you can paste the &lt;strong&gt;TikTok profile&lt;/strong&gt; or &lt;strong&gt;YouTube channel&lt;/strong&gt; link directly, set &lt;strong&gt;Max videos per URL&lt;/strong&gt; (for example 50) and optionally &lt;strong&gt;Only videos published after&lt;/strong&gt; a date.&lt;/p&gt;

&lt;p&gt;Transcribing a whole niche is a two-step pipeline:&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;
&lt;strong&gt;Collect video URLs&lt;/strong&gt; with a scraper such as &lt;a href="https://apify.com/clockworks/tiktok-scraper" rel="noopener noreferrer"&gt;TikTok Scraper&lt;/a&gt; (hashtags, searches) or &lt;a href="https://apify.com/apify/instagram-reel-scraper" rel="noopener noreferrer"&gt;Instagram Reel Scraper&lt;/a&gt; (Instagram only lists a profile's posts to logged-in users, so profiles have to go through a scraper).&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Pass the URLs&lt;/strong&gt; into Video to Text with &lt;strong&gt;AI analysis&lt;/strong&gt; turned on.&lt;/li&gt;
&lt;/ol&gt;

&lt;p&gt;Every row then already has its hook, key points and call to action. Feed them to an LLM with a prompt like:&lt;/p&gt;

&lt;blockquote&gt;
&lt;p&gt;"Here are 200 transcripts of top-performing TikToks about home workouts. List the 10 most common hooks in the first 3 seconds, with an example of each."&lt;/p&gt;
&lt;/blockquote&gt;

&lt;p&gt;You get a research report on what actually works in a niche in minutes instead of days of watching videos.&lt;/p&gt;

&lt;h2&gt;
  
  
  Using it from an AI agent
&lt;/h2&gt;

&lt;p&gt;The Actor is available through the &lt;strong&gt;Apify MCP server&lt;/strong&gt;, so an agent in Claude, ChatGPT or Cursor can transcribe videos on its own: give it a link, and it gets back the transcript as JSON.&lt;/p&gt;

&lt;h2&gt;
  
  
  How much does it cost?
&lt;/h2&gt;

&lt;p&gt;You pay only for successful transcripts: &lt;strong&gt;$0.006 per video&lt;/strong&gt; up to 2 minutes (about &lt;strong&gt;$6 per 1,000 Reels or TikToks&lt;/strong&gt;), plus $0.003 for each extra minute of AI transcription. The add-ons are optional: on-screen text costs $0.002, AI analysis $0.004 per video and translation $0.003 per 1,000 words. Failed videos are free, and the Apify free plan's monthly credit covers roughly 800 short videos.&lt;/p&gt;

&lt;p&gt;Working with podcasts instead? The companion &lt;a href="https://apify.com/aemon/podcast-audio-to-text" rel="noopener noreferrer"&gt;Podcast to Text&lt;/a&gt; Actor takes Apple Podcasts links and RSS feeds and charges per minute of audio.&lt;/p&gt;

&lt;h2&gt;
  
  
  Wrapping up
&lt;/h2&gt;

&lt;p&gt;Turning short-form video into text unlocks it for search, analysis and AI. Use platform captions when they exist, Whisper when they don't, and keep the output structured so it plugs straight into your pipeline.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;What would you build with transcripts of thousands of short videos?&lt;/strong&gt; I'd love to hear your use cases in the comments. And if a link doesn't work, open an issue on the Actor page.&lt;/p&gt;

&lt;p&gt;&lt;em&gt;Disclosure: I'm the author of the Video to Text Actor used in the examples, published on Apify under my agency &lt;a href="https://www.aemon.cz/" rel="noopener noreferrer"&gt;Aemon&lt;/a&gt;.&lt;/em&gt;&lt;/p&gt;

</description>
      <category>ai</category>
      <category>python</category>
      <category>javascript</category>
      <category>tutorial</category>
    </item>
  </channel>
</rss>
