Pricing

from $0.50 / 1,000 videos

Try for free

Go to Apify Store

YouTube Transcript Scraper - Any Video or Whole Channel

Try for free

Transcripts for one YouTube video or a whole channel, with views, dates, comments and Shorts in the same run - one Actor, not three. No API key, no quota. Exact counts, not 3.1M. Cached, failed and filtered rows are free, so re-runs cost almost nothing.

Pricing

from $0.50 / 1,000 videos

Rating

0.0

(0)

Developer

Lorenzo

Actor stats

Bookmarked

Total users

Monthly active users

18 days ago

Last modified

Categories

Developer tools

Videos

Social media

Channels, playlists, videos or search terms

inputs

Required

One per line, mixed freely. Channel: youtube.com/@handle, @handle, or a /channel/UC... URL — returns that channel's videos, newest first. Playlist: a playlist URL or a bare list ID. Video: an 11-character video ID, a watch URL, youtu.be, /shorts/ or /embed/. Search: any words, e.g. drone review. Note: a bare 11-character word is read as a video ID (so is 'helicopters') — write search:helicopters to force a search.

Type:array

Min. items:1

Max videos per channel, playlist or search

maxResultsPerSource

Optional

How many videos to take from each channel, playlist or search term. Videos passed by ID ignore this. This is the main cost control: 10 channels at 100 each is 1,000 videos. Playlists return 100 IDs per request, so values that are multiples of 100 are the cheapest per video.

Type:integer

Minimum:1

Maximum:4000

Default:30

Which channel tabs to read

channelTabs

Optional

A channel keeps long videos, Shorts and live streams in three separate tabs that do not overlap — measured on one channel: 30 videos, 48 Shorts, 5 streams. All three are read by default so "every video from the channel" means every video. Each tab costs one extra listing request; drop the ones you do not want. Every row says which tab it came from in video_type.

Type:string[]

Default:

[
  "videos",
  "shorts",
  "streams"
]

Published on or after (YYYY-MM-DD)

publishedAfter

Optional

Keep only videos published on or after this date. Leave empty for no limit. Filtering happens after each video is read, so videos that are filtered out are never charged to you.

Type:string

Published on or before (YYYY-MM-DD)

publishedBefore

Optional

Keep only videos published on or before this date. Leave empty for no limit.

Type:string

Only videos from the last N days (rolling)

publishedWithinDays

Optional

For scheduled runs. Keeps only videos published in the last N days, counted from the day the run starts, so a saved task keeps returning what is NEW instead of the same window forever. 0 or empty is off. An explicit 'Published on or after' date wins if you set both. Videos you already fetched come back from stored state and are not charged, so a daily or weekly watch costs a fraction of the first run.

Type:integer

Minimum:0

Maximum:3650

Default:0

Minimum duration (seconds)

durationMinSeconds

Optional

Keep only videos at least this long. 0 means no minimum. Use 61 to exclude Shorts.

Type:integer

Minimum:0

Default:0

Maximum duration (seconds)

durationMaxSeconds

Optional

Keep only videos no longer than this. 0 means no maximum.

Type:integer

Minimum:0

Default:0

Include like counts

includeLikes

Optional

Off by default. Adds a like count to every row from a small extra request (about 1,392 bytes). Not charged separately.

Type:boolean

Default:false

Include transcripts / subtitles (billed separately)

includeTranscript

Optional

Off by default. Adds the full transcript text, its language, whether it was auto-generated, and the number of timed cues. Getting subtitles needs a different and larger request per video, so it is charged as its own event and only when a transcript is actually returned. Videos with subtitles disabled come back with transcript_status = NO_CAPTIONS and cost you nothing.

Type:boolean

Default:false

Preferred transcript languages

transcriptLanguages

Optional

In order of preference, e.g. en, es, de. The first available match wins. If none match, the video's own first track is used — human-written subtitles before auto-generated ones. Only used when transcripts are on.

Type:array

Default:

[
  "en"
]

Add timed transcript segments (no extra charge)

includeTranscriptSegments

Optional

Off by default, and only used when transcripts are on. Adds transcript_segments: one entry per subtitle cue with start_s, duration_s and text — the same words as transcript, split by timestamp. Turn it on when you need to cite a moment in the video. Leave it off for text: a single video can carry 1,000 cues, every export format flattens them into transcript_segments/0/text … /999/text, and a spreadsheet of a few videos becomes thousands of columns. It costs nothing extra either way — the cues arrive in the same request as the text — and transcript_cue_count tells you how many there would be.

Type:boolean

Default:false

Include comments (billed separately)

includeComments

Optional

Off by default. Adds comments as EXTRA ROWS, not extra columns — each comment is its own row with _row_type = comment, which is what you want in a CSV. Video rows are unchanged and keep _row_type = video. Charged per comment returned. Videos with comments turned off cost you nothing.

Type:boolean

Default:false

Max comments per video

maxCommentsPerVideo

Optional

Top-level comments come 20 per request, so 20, 40 or 100 are natural stopping points. Only used when comments are on.

Type:integer

Minimum:1

Maximum:1000

Default:20

Replies to fetch per comment

maxRepliesPerComment

Optional

0 = top-level comments only. Replies need one extra request per comment that has any, so this is capped per comment rather than priced per reply — it stops 100 comments quietly becoming 100 extra requests. Replies arrive as their own rows with comment_is_reply = true.

Type:integer

Minimum:0

Maximum:100

Default:0

Serve from cache if fetched within (hours)

maxAgeHours

Optional

Videos this actor already fetched within this many hours are returned from stored state instead of being fetched again, and are not charged. Cached rows are marked _cached = true with _cached_at, so row age is always visible. Set 0 to force a fresh fetch. View counts move constantly — use 0 when sorting on views or computing growth.

Type:integer

Minimum:0

Maximum:720

Default:24

Reset the drift baseline on this run

resetFingerprintBaseline

Optional

Advanced. Every run compares its column shape against a stored baseline and warns when fields start coming back empty. After an intended change to the output columns the old baseline is permanently wrong and warns forever. Turn this on for one run to replace it. Leave off otherwise — a baseline that resets itself cannot detect anything.

Type:boolean

Default:false

Parallel requests

concurrency

Optional

How many requests are in flight at once. Each worker makes at most one request per second, so this also sets the request rate. Lower it if you see failures; raising it finishes sooner but does not cost less.

Type:integer

Minimum:1

Maximum:16

Default:8

Language (hl)

language

Optional

Language for text YouTube generates — category names, "N subscribers", relative dates. e.g. en, de, es, ja. Titles and descriptions are whatever the creator wrote and do not change.

Type:string

Default:en

Region (gl)

region

Optional

Two-letter country the results are ranked and filtered for, e.g. US, DE, GB, IN. Mostly affects search ordering and availability.

Type:string

Default:US

Max videos in total (0 = no limit)

maxItems

Optional

Hard cap across the whole run, applied after channels and playlists are expanded. 0 processes everything.

Type:integer

Minimum:0

Default:0

Proxy — RESIDENTIAL required

proxyConfiguration

Optional

RESIDENTIAL is required and is the default. Measured on this actor: channel and playlist listings work on datacenter IPs, but the per-video request was refused on 20 of 20 with "Sign in to confirm you're not a bot". Responses are small — about 4.4 KB per video plus one listing request per channel or playlist — so bandwidth is roughly $0.0001 per video at $8/GB.

Type:object

Default:

{
  "useApifyProxy": true,
  "apifyProxyGroups": [
    "RESIDENTIAL"
  ]
}

Site

site

Optional

Which site config to use. Only 'youtube' today.

Type:string

Default:youtube

Options:

youtube

Youtube Free Playlist Scraper

scrapesmith/youtube-free-playlist-scraper

⚡ Free Fast YouTube Playlist Scraper – Extract all videos from any Playlist with titles, views, likes, comments, publish dates, thumbnails & subscriber count. No proxies needed. Get thousands of channel videos in minutes, 100% free.

Scrape Smith

YouTube Scraper — Videos, Transcripts, Comments, Channels

hipersoft/youtube-scraper

All-in-one YouTube API: get video metadata, full transcripts, comments with replies, channel data, and search results in a single call. No setup, fast and cheap.

hiper soft

YouTube Full Channel Transcripts Extractor

karamelo/youtube-full-channel-transcripts-extractor

With only the channel or playlist link You can extract 1 to 1000s of all the transcripts of a channel, be it videos or shorts or streams/lives or even podcasts and playlists, you name it. Get all the transcripts/captions organized with video ID and title in a nice table or JSON or CSV to download.

karamelo

4.9

Youtube Transcript Scraper - $0.7 per 1k

supreme_coder/youtube-transcript-scraper

Scrape youtube videos and transcripts in bulk from channels, youtube search, playlists, etc. Provides videos details and transcript in different formats

Supreme Coder

2.1K

5.0

Fast YouTube Playlist Scraper API | Extract Videos & Metadata

apidojo/youtube-playlist-scraper

The ultimate solution for detailed YouTube playlist information. Enjoy unmatched speed and thoroughness in both search and direct video retrieval from playlists. Additionally, it's remarkably cost-effective at just $0.50 per 1000 videos!

API Dojo

583

5.0

YouTube Channel Transcripts — Bulk Extract by Influship

influship/youtube-channel-transcripts

Extract transcripts from all recent videos on a YouTube channel in a single run. Returns full text with timestamps per video. Ideal for content research and AI training. No login required.

Influship

206

5.0

YouTube Transcript & Subtitle Scraper

abotapi/youtube-transcript-scraper

Extract transcripts and subtitles from YouTube videos in bulk using video, playlist, channel URLs, or keyword search. Returns timed transcript segments, plain text, SRT, and WebVTT subtitle files, with optional auto-translation to other languages.

Abot API

🎬 Youtube Playlist Extractor

dz_omar/youtube-playlist-extractor

Extract complete 🎬 playlist information with all video details from any YouTube playlist -->⚡Fast, reliable, and built for scale. Get video lists, durations, thumbnails, and channel info.

FlowExtract API

412

4.9

YouTube Transcript Master [EASY] (WebVTT & more)

zerohour/yt-transcript

YouTube Transcripts in BULK! Easily query via channel, playlist, or video URLs. Built with simplicity & reliability in mind, with expert support. Perfect data to feed your AI or LLM. Output multiple formats: TEXT, JSON, SRV, TTML, VTT (WebVTT). Automatic YouTube captions are available as backup.

Zerohour

131

Fast YouTube Channel Scraper

streamers/youtube-channel-scraper

This alternative YouTube Data API has no limits or quotas. Use it to scrape one or multiple YouTube channels: channel info, URL, total number of subscribers, videos and views, creation date. Try it and get basic video data. You can download extracted data in JSON, CSV, and Excel.