YouTube Search Transcripts: transcripts for any search, one run

Search YouTube and get the transcript of every result in one run, up to 50 queries: one row per video with rank, title, channel, views and the full transcript text. Videos without captions cost nothing. Pay per transcript.

Why this scraper

• Only delivered results are charged. Inputs that fail come back as clear error records at no cost.
• One run, no glue: the search runs over the slim Innertube route (about 2 KB per result) and every result goes straight into the transcript route. Measured on "how to make sourdough bread": 5 results, 5 transcripts of 8,000 to 18,000 characters, in a single run.
• The transcript is in the language the video is spoken in. When a video carries community translations, the spoken language wins over the first translation in the list; pass languages to ask for a specific one. A result without captions is never a row and never charged; it is counted in skippedNoCaptions.

Who this is for

Type your queries in queries (up to 50 per run), set maxResultsPerQuery (default 10, ceiling 100) and, if you want, sortBy (relevance, upload date, view count or rating). Every delivered row carries the query, the rank in the search results, the video's title, views and age, the transcript language, whether the captions were auto-generated and the full transcript text. Turn on includeTimestampedSnippets when the text has to be aligned to the video.

Who this is not for

Only videos that have captions (manual or auto-generated) produce a row; music, very new uploads and some live streams have none and are counted in skippedNoCaptions instead. Search results are YouTube's, personalised for nobody but still varying a little between runs and regions, so the same query can return a slightly different set tomorrow. The transcript is text, not a translation: you get the captions in the video's spoken language unless you list preferred languages.

Input fields

Field	Type	Required or default	What it does
queries	list of text	required	One per row, up to 50, exactly as you would type it in YouTube search.
maxResultsPerQuery	number	10	Cost ceiling per query, in search order.
sortBy	text (relevance, upload_date, view_count, rating)	relevance	YouTube's own sort of the results.
languages	list of text	empty	Language codes in order of preference, for example en, nl, de. Empty takes the video's default track.
preferManualCaptions	true/false	true	On, a human-made track wins over an auto-generated one when both exist.
includeTimestampedSnippets	true/false	false	On, each row also carries the transcript as timed snippets.

Input example

{
    "queries": [
        "how to make sourdough bread"
    ],
    "maxResultsPerQuery": 10,
    "sortBy": "relevance",
    "languages": [],
    "preferManualCaptions": true,
    "includeTimestampedSnippets": false
}

Output example

Field	Type	What it holds	Example
query	text	The search term this row was found with.	how to make sourdough bread
sortBy	text	The ranking that was asked for: relevance, date, views or rating.	relevance
rank	number	The place of the video in the search results, counting from 1.	1
videoId	text	The eleven-character YouTube id of the video, the key to join other data on.	4gEoh3sk2AE
url	text	The watch page of the video.	https://www.youtube.com/watch?v=4gEoh3sk2AE
title	text	The title as the uploader set it.	Bake the Perfect Sourdough Bread: A Step-by-Step Guide
viewCount	number	Views as YouTube reports them at the moment of the run.	3304701
publishedText	text	How long ago YouTube says it appeared, for example "3 weeks ago"; YouTube gives no exact date here.	2 years ago
durationSeconds	number	The length of the video in seconds.	712
language	text	The caption language the transcript was taken from, as a language code.	ar
isAutoGenerated	true/false	True when the captions come from YouTube's speech recognition instead of the channel.	False
text	text	The full transcript of the video as one block of text.	- سأوضح لك اليوم كيفية صنع الخبز المخمر. إليك كل ما تحتاج...
snippets	list	The same transcript in timed pieces, each with its start time in seconds and its text.
skippedNoCaptions	number	How many videos in this input were passed over because they carry no captions at all.	1
skippedBlocked	number

Error codes: INVALID_QUERY, INVALID_SORT, NO_RESULTS, NO_TRANSCRIPT, BLOCKED.

One delivered row looks like this:

{
  "query": "how to make sourdough bread",
  "sortBy": "relevance",
  "rank": 1,
  "videoId": "4gEoh3sk2AE",
  "url": "https://www.youtube.com/watch?v=4gEoh3sk2AE",
  "title": "Bake the Perfect Sourdough Bread: A Step-by-Step Guide",
  "viewCount": 3241316,
  "publishedText": "2 years ago",
  "durationSeconds": 712,
  "language": "en",
  "isAutoGenerated": false,
  "text": "- Today I am gonna show you how to make sourdough bread. Here's everything you need to know to get the beautiful rise, the perfect little ear, and the best text...",
  "skippedNoCaptions": 0,
  "status": "ok"
}

Related actors from steadydata

• youtube-search: the same search results without transcripts
• youtube-channel-transcripts: the same idea for a whole channel
• youtube-transcript-bulk: transcripts for a list of specific videos

Pricing

Pay per event: one transcript-fetched event per delivered result. No charge for inputs that fail, no separate platform-usage surcharge.

Free Apify plan: this actor delivers up to 25 rows per run for accounts on the Apify free plan, and then stops with a message. That limit is set by us, not by Apify. It exists so the actor keeps paying for itself for the people who do pay. Any paid Apify plan runs it at full size, billed per delivered row, with failed rows never charged.

Reviews: if this actor saves you time, a short review on this page is the one thing that helps most. Ratings are what other buyers look at first, and we have no other way to ask.

Use with AI agents

AI assistants such as Claude, ChatGPT or Cursor can run this actor through Apify's own MCP server. Ask for the actor steadydata/youtube-search-transcripts by its full name, so the assistant does not pick a different one, and start from this minimal input:

{
  "queries": [
    "how to make sourdough bread"
  ]
}

The assistant can read the price per event from the Pricing section above before the first paid run. Failed input comes back as an error row, as described above.

FAQ

Do I pay for results without captions? No. Only a delivered transcript is charged; a query that fails or a video without captions costs nothing.

How many results can I take per query? Up to 100 (maxResultsPerQuery), in the order YouTube returns them for the chosen sortBy.

Which language do I get? The captions in the language the video is spoken in: a manual track when it exists, otherwise the auto-generated one. Pass languages (for example ["nl", "en"]) to prefer specific tracks, and preferManualCaptions: false to take auto-generated captions first.

Can I get timestamps? Yes, set includeTimestampedSnippets to true and every row gets a snippets list with start time, duration and text per caption line.

Can I search a channel or a playlist instead? Use YouTube Channel Transcripts or YouTube Playlist Transcripts from the related actors below; this actor is for open search queries.

Is personal data collected? The transcript is the caption track YouTube exposes for that video; it is not generated by this actor and the text follows the captioner, including its punctuation and line breaks.

What happens when the source changes? Sources change from time to time; that is the nature of this work. The actor is monitored daily and fixed fast, and while it is broken you are not charged, because only delivered results cost anything.

Can I call it from my own code, like an API? Yes. Every run is available through the Apify API: start it with one HTTP request, then read the results as JSON, CSV or Excel from the dataset endpoint, or use the official Python and JavaScript clients. The input is the same JSON as in the input example above, so a scheduled script, a Make or Zapier step or an AI agent can call it without any changes.