Podcast Scraper – Episodes & Transcripts

sfoln/podcast-scraper · AI, News, Automation On request

Get any podcast's episodes as JSON from its name, Apple Podcasts link or RSS feed: titles, dates, durations, audio URLs, show notes, episode and season numbers, plus a show profile with release cadence. Optional AI transcripts. $1 per 1,000 episodes.

Request access Run it$1 / 1,000 podcast episode, more below

About

What it does

Get any podcast's episodes as clean JSON from its name, an Apple Podcasts link or its RSS feed. For every show you get one row per episode (title, publication date, duration, audio file URL, show notes as text, the links in them, episode and season numbers, episode type, explicit flag, GUID) and one show profile with the owner's published contact, categories and release cadence. Optionally, the newest episodes are transcribed with AI (Mistral Voxtral). Built for podcast research, PR and guest booking, ad sales, media monitoring and AI agents that need what was said on a show. $1 per 1,000 episodes; transcripts $0.0105 per audio minute.

Price

Price per delivered record

Event	Price	Per 1,000	Charged for
Podcast episode `episode`	$0.001	$1 / 1,000	Charged once per podcast episode delivered (title, date, duration, audio URL, show notes, episode and season numbers).
Transcribed audio minute `transcription-minute` paid plans only	$0.0105	$10.5 / 1,000	Charged per started minute of audio transcribed for the transcription add-on (a 32 min 10 s episode = 33 events). Paid plans only.

Apify free plan: up to 10 items per run. Paid plans have no per-run cap; set maxTotalChargeUsd to cap spend.

Run

Call it from code, an agent or the Console

This API is private on Apify for now. Email us and we will publish it for you, usually within two working days. The calls below work once it is public. contact@sfoln.com

curl · synchronous run, rows in the response

curl -X POST "https://api.apify.com/v2/actors/sfoln~podcast-scraper/run-sync-get-dataset-items" \
  -H "Authorization: Bearer $APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"podcasts":["https://feeds.npr.org/510289/podcast.xml"],"maxEpisodesPerShow":5}'

mcp.json · hosted Apify MCP server

{
  "mcpServers": {
    "apify": {
      "type": "http",
      "url": "https://mcp.apify.com?tools=sfoln/podcast-scraper"
    }
  }
}

asynchronous run · for large batches

POST https://api.apify.com/v2/actors/sfoln~podcast-scraper/runs?maxTotalChargeUsd=1
GET  https://api.apify.com/v2/actor-runs/{runId}            # until status is SUCCEEDED
GET  https://api.apify.com/v2/datasets/{defaultDatasetId}/items?format=json

Authenticate with an Apify API token. MCP clients can instead sign in with OAuth when prompted. Example prompt for an agent: List the 5 latest episodes of the Planet Money podcast with their dates, durations and audio links.

Input

Input fields

Podcasts to read: show names, Apple Podcasts links or ids, RSS feed URLs or podcast websites. Each podcast gives one row per episode plus one show profile row, and optionally transcript rows.

Field	Type	Default	Meaning
`podcasts` *	`array`	required	Podcasts One podcast per line: a show name (the best Apple Podcasts match is used), an Apple Podcasts link or id, an RSS feed URL, or the show's website. Accounts on the Apify free plan can process up to 10 podcasts per run (20 episodes each).
`maxEpisodesPerShow`	`integer`	`50`	Max episodes per show Newest episodes to return per show (1–5,000). Range 1–5000.
`sinceDays`	`integer`	`0`	Only episodes from the last N days Keep only episodes published in the last N days (0 = all episodes, up to the maximum above). Range 0–36500.
`country`	`string`	`"us"`	Apple Podcasts country Two-letter storefront for show names and Apple ids, e.g. us, gb, de (an Apple Podcasts link's own country is used for that link).
`transcribeLatest`	`integer`	`0`	Transcribe the latest N episodes Paid add-on: transcribe the newest N delivered episodes of each show with Mistral Voxtral ($0.0105 per audio minute, paid Apify plans only). Before the audio is sent, the remaining maximum cost per run must cover the longest the audio file can last (its real size at 32 kbps, at most 3 hours); only the minutes transcribed are charged. 0 = no transcription. Range 0–10.
`maxMinutesPerEpisode`	`integer`	`90`	Max minutes per transcribed episode Episodes longer than this (by the feed's duration, else the audio file's size) are not transcribed and not charged (1–180). Range 1–180.
`transcriptionLanguage`	`string`	—	Transcription language ISO 639-1 code of the spoken language (e.g. en, de, es) to help the transcription; empty = detected automatically.
`timestamps`	`boolean`	`false`	Transcript timestamps Add timed segments ({ start, end, text }) to each transcript.
`maxShows`	`integer`	`100`	Max podcasts Safety cap on how many podcasts are processed in this run; the rest are ignored and counted in the OUTPUT record. Range 1–1000.
`concurrency`	`integer`	`3`	Concurrency Podcasts processed in parallel (1–5). Apple's directory is asked at most about 20 times per minute per run. Range 1–5.
`includeRaw`	`boolean`	`false`	Include raw data Attach each episode's parsed feed item (all namespaces) and the Apple directory record of the show (larger output).

Example input

{
  "podcasts": [
    "https://feeds.npr.org/510289/podcast.xml"
  ],
  "maxEpisodesPerShow": 5
}

Output

One record per input item

Records land in the run's dataset; a synchronous call returns them directly. Each has a status (ok is charged; everything else is free) and a charged flag. The run's OUTPUT record totals items and charges.

Field	Type	Meaning
`input`	`string \| object`	Input The podcast this record belongs to, as given.
`recordType`	`string`	Record type episode = one episode (charged); transcript = one transcribed episode (charged per audio minute); summary = the show profile, one per podcast (free). e.g. `"episode"`
`position`	`integer`	Position episode: 1-based position among the episodes delivered for the podcast (newest first).
`showTitle`	`string \| null`	Show The podcast's title.
`feedUrl`	`string \| null`	Feed URL The show's RSS feed (after redirects).
`appleId`	`string \| null`	Apple id The show's Apple Podcasts id, when it was looked up there.
`guid`	`string \| null`	GUID episode / transcript: the episode's GUID in the feed.
`title`	`string \| null`	Episode title episode: the episode's title.
`publishedAt`	`string \| null`	Published at episode / transcript: ISO 8601 publication time.
`durationSecs`	`number \| null`	Duration (s) episode: length in seconds as the feed states it; transcript: the audio length Mistral reports.
`duration`	`string \| null`	Duration episode: length as h:mm:ss.
`audioUrl`	`string \| null`	Audio URL episode / transcript: the audio file (the feed's enclosure).
`audioType`	`string \| null`	Audio type episode: MIME type of the audio file.
`audioSizeBytes`	`integer \| null`	Audio size episode: size of the audio file in bytes, as the feed states it.
`episodeNumber`	`integer \| null`	Episode number episode: itunes:episode (or podcast:episode).
`seasonNumber`	`integer \| null`	Season number episode: itunes:season (or podcast:season).
`episodeType`	`string \| null`	Episode type episode: full \| trailer \| bonus.
`explicit`	`boolean \| null`	Explicit Explicit content flag (an episode without its own flag inherits the show's).
`link`	`string \| null`	Episode page episode: the episode's web page.
`imageUrl`	`string \| null`	Image episode: the episode artwork; summary: the show artwork.
`description`	`string \| null`	Description episode: the show notes as plain text (at most 20,000 characters); summary: the show's description.
`links`	`array \| null`	Show-note links episode: the links in the show notes (at most 50).
`transcripts`	`array \| null`	Published transcripts episode: transcripts the show publishes itself (podcast:transcript) as [{ url, type, language }].
`chaptersUrl`	`string \| null`	Chapters episode: the podcast:chapters file, when the show publishes one.
`people`	`array \| null`	People episode: hosts and guests the feed credits (podcast:person) as [{ name, role, group }].
`appleEpisodeId`	`string \| null`	Apple episode id episode: Apple's episode id (only for episodes taken from Apple's directory).
`appleEpisodeUrl`	`string \| null`	Apple episode URL episode: the Apple Podcasts page of the episode (only for episodes taken from Apple's directory).
`episodeTitle`	`string \| null`	Episode title transcript: the transcribed episode's title.
`text`	`string \| null`	Transcript transcript: the transcribed text.
`segments`	`array \| null`	Segments transcript: timed segments [{ start, end, text, speaker }] (with timestamps).
`billedMinutes`	`integer \| null`	Billed minutes transcript: started minutes of audio charged (transcription-minute events).
`wordCount`	`integer \| null`	Words transcript: words of the transcript.
`model`	`string \| null`	Model transcript: the transcription model.
`podcast`	`string \| null`	Podcast summary: the podcast as given.
`resolvedVia`	`string \| null`	Resolved via summary: search \| apple_id \| feed_url \| website.
`episodesSource`	`string \| null`	Episodes from summary: feed (the show's RSS feed) \| apple (Apple's directory, when the show has no readable feed).
`author`	`string \| null`	Author summary: itunes:author, else Apple's artist name.
`ownerName`	`string \| null`	Owner name summary: itunes:owner name as published in the feed.
`ownerEmail`	`string \| null`	Owner e-mail summary: itunes:owner e-mail as published in the feed.
`language`	`string \| null`	Language summary: the feed's language; transcript: the detected spoken language.
`categories`	`array \| null`	Categories summary: iTunes categories ("Business > Investing").
`appleGenres`	`array \| null`	Apple genres summary: Apple's genres for the show.
`showType`	`string \| null`	Show type summary: episodic \| serial.
`complete`	`boolean \| null`	Complete summary: the show has ended (itunes:complete).
`website`	`string \| null`	Website summary: the show's website (the feed's link).
`newFeedUrl`	`string \| null`	New feed URL summary: the feed the show moved to (itunes:new-feed-url).
`copyright`	`string \| null`	Copyright summary: the feed's copyright line.
`podcastGuid`	`string \| null`	Podcast GUID summary: Podcasting 2.0 podcast:guid.
`fundingUrls`	`array \| null`	Funding links summary: podcast:funding links.
`appleUrl`	`string \| null`	Apple Podcasts URL summary: the show's Apple Podcasts page.
`appleEpisodeCount`	`integer \| null`	Apple episode count summary: episodes Apple lists for the show.
`episodesInFeed`	`integer \| null`	Episodes in feed summary: episodes the feed (or Apple) lists.
`firstEpisodeAt`	`string \| null`	First episode summary: date of the oldest listed episode.
`latestEpisodeAt`	`string \| null`	Latest episode summary: date of the newest episode.
`averageDurationSecs`	`integer \| null`	Average duration (s) summary: average episode length.
`medianIntervalDays`	`number \| null`	Release cadence (days) summary: median days between the latest 20 episodes.
`episodesLast30Days`	`integer \| null`	Episodes, last 30 days summary: episodes published in the last 30 days.
`episodesLast90Days`	`integer \| null`	Episodes, last 90 days summary: episodes published in the last 90 days.
`episodesOutsideDateWindow`	`integer \| null`	Episodes outside sinceDays summary: episodes left out by sinceDays (not charged).
`otherMatches`	`array \| null`	Other matches summary (name search): other Apple Podcasts matches [{ appleId, title, author, feedUrl, episodeCount }]; give one's appleId to read it instead.
`transcription`	`object \| null`	Transcription summary: { requested, delivered, minutesBilled, episodes: [{ guid, title, status, minutes?, reason? }] } when transcribeLatest is set on a paid plan.
`resultsDelivered`	`integer \| null`	Episodes delivered summary: episode rows delivered (and charged) for this podcast.
`stopReason`	`string \| null`	Stop reason summary: limit (maxEpisodesPerShow reached) \| end (no more episodes) \| budget \| free_plan \| aborted.
`status`	`string`	Status ok \| not_found \| error \| skipped_invalid \| skipped_budget \| skipped_free_plan \| skipped_unmetered e.g. `"ok"`
`charged`	`boolean`	Charged episode and transcript rows: true (charged before they were written); summary rows: false.
`chargedEvent`	`string`	Charged event episode on episode rows; transcription-minute on transcript rows (billedMinutes events).
`chargedCount`	`integer`	Events charged Events actually charged when a charge came back short (present only then).
`source`	`string`	Source podcast: the show's RSS feed and Apple's podcast directory (transcripts by Mistral Voxtral).
`fetchedAt`	`string`	Fetched at ISO timestamp.
`error`	`object`	Error { code, message, provider?, status? } when status is not ok: invalid_input, podcast_not_found, not_a_feed, malformed_feed, not_found, http_error, rate_limit (Apple throttling), timeout, network, blocked_url; runtime codes: budget_exhausted, free_plan_limit, unmetered, aborted, charge_failed, record_too_large, invalid_record, dataset_push_failed.
`warnings`	`array`	Warnings Notes about this record (caps, Apple fallback, transcripts not included and why).
`raw`	`object \| null`	Raw data episode: the parsed feed item; summary: Apple's show record (only with includeRaw).

Example record

[
  {
    "input": "https://feeds.npr.org/510289/podcast.xml",
    "recordType": "episode",
    "position": 1,
    "showTitle": "Planet Money",
    "feedUrl": "https://feeds.npr.org/510289/podcast.xml",
    "appleId": null,
    "guid": "b297c28a-87e4-4b96-92e0-e64b7b581948",
    "title": "Wanna slow down AI? These biologists can help.",
    "publishedAt": "2026-10-14T21:30:18.000Z",
    "durationSecs": 1750,
    "duration": "29:10",
    "audioUrl": "https://play.podtrac.com/npr-510289/npr.simplecastaudio.com/episodes/fe8692b3/audio/128/default.mp3",
    "audioType": "audio/mpeg",
    "audioSizeBytes": 28012130,
    "episodeNumber": null,
    "seasonNumber": null,
    "episodeType": "full",
    "explicit": false,
    "link": "https://www.npr.org/2026/10/14/nx-s1-6002910/ai-scientific-regulation-conference-asilomar",
    "imageUrl": "https://npr.brightspotcdn.com/dims3/default/strip/false/crop/3377x1900+0+0/resize/1280/quality/66/format/jpg/",
    "description": "Fifty years ago, scientists pioneering a terrifying gene-splicing technology figured out a way to collectively hit pause. Can we do it again with AI? …",
    "links": [
      "https://kilkenomics.com/",
      "https://www.npr.org/newsletter/indicator"
    ],
    "transcripts": [],
    "chaptersUrl": null,
    "people": [],
    "appleEpisodeId": null,
    "appleEpisodeUrl": null,
    "status": "ok",
    "charged": true,
    "chargedEvent": "episode",
    "source": "podcast",
    "fetchedAt": "2026-10-15T09:00:02.000Z"
  },
  {
    "input": "https://feeds.npr.org/510289/podcast.xml",
    "recordType": "transcript",
    "guid": "b297c28a-87e4-4b96-92e0-e64b7b581948",
    "episodeTitle": "Wanna slow down AI? These biologists can help.",
    "showTitle": "Planet Money",
    "publishedAt": "2026-10-14T21:30:18.000Z",
    "audioUrl": "https://play.podtrac.com/npr-510289/npr.simplecastaudio.com/episodes/fe8692b3/audio/128/default.mp3",
    "language": "en",
    "text": "This is Planet Money from NPR. …",
    "durationSecs": 1750.4,
    "billedMinutes": 30,
    "wordCount": 4630,
    "model": "voxtral-mini-2602",
    "status": "ok",
    "charged": true,
    "chargedEvent": "transcription-minute",
    "source": "podcast",
    "fetchedAt": "2026-10-15T09:01:10.000Z"
  },
  {
    "input": "https://feeds.npr.org/510289/podcast.xml",
    "podcast": "https://feeds.npr.org/510289/podcast.xml",
    "resolvedVia": "feed_url",
    "episodesSource": "feed",
    "showTitle": "Planet Money",
    "author": "NPR",
    "ownerName": "NPR",
    "ownerEmail": "podcasts@npr.org",
    "description": "Wanna see a trick? Give us any topic and we can tie it back to the economy. …",
    "language": "en",
    "categories": [
      "Business",
      "News"
    ],
    "appleGenres": [],
    "explicit": false,
    "showType": "episodic",
    "complete": null,
    "website": "https://www.npr.org/podcasts/510289/planet-money",
    "feedUrl": "https://feeds.npr.org/510289/podcast.xml",
    "newFeedUrl": null,
    "imageUrl": "https://media.npr.org/assets/img/2022/10/24/pm_new_tile_2022_sq.jpg",
    "copyright": "Copyright 2015-2026 NPR - For Personal Use Only",
    "podcastGuid": null,
    "fundingUrls": [
      "https://plus.npr.org/"
    ],
    "appleId": null,
    "appleUrl": null,
    "appleEpisodeCount": null,
    "episodesInFeed": 355,
    "firstEpisodeAt": "2023-06-16T18:28:34.000Z",
    "latestEpisodeAt": "2026-10-14T21:30:18.000Z",
    "averageDurationSecs": 1635,
    "medianIntervalDays": 2.9,
    "episodesLast30Days": 9,
    "episodesLast90Days": 28,
    "episodesOutsideDateWindow": 0,
    "transcription": {
      "requested": 1,
      "delivered": 1,
      "minutesBilled": 30,
      "episodes": [
        {
          "guid": "b297c28a-87e4-4b96-92e0-e64b7b581948",
          "title": "Wanna slow down AI? These biologists can help.",
          "status": "delivered",
          "minutes": 30
        }
      ]
    },
    "recordType": "summary",
    "resultsDelivered": 20,
    "stopReason": "limit",
    "status": "ok",
    "charged": false,
    "source": "podcast",
    "fetchedAt": "2026-10-15T09:01:10.000Z"
  }
]

FAQ

Questions

What can I give as a podcast?: A show name or search term (the best match in Apple Podcasts is used, other matches are listed in the summary), an Apple Podcasts link or id (https://podcasts.apple.com/us/podcast/…/id290783428, or 290783428), an RSS feed URL, or the show's website (its feed is found from the page). Spotify links have no public feed, so give the show's name instead.
Where does the data come from?: From the show's own RSS feed, which every podcast publishes for directories: episode titles, dates, durations, audio files, show notes, iTunes episode and season numbers and Podcasting 2.0 tags (transcript and chapter links, people). The show is looked up in Apple's public podcast directory to find the feed. Shows without a public feed (some Apple-exclusive shows) get Apple's latest 200 episodes with fewer fields.
How does the transcription add-on work?: Set transcribeLatest to transcribe the newest delivered episodes of each show with Mistral Voxtral (paid Apify plans). Each transcript is a separate row (recordType transcript) and costs $0.0105 per started minute of audio. Episodes longer than maxMinutesPerEpisode, or whose audio file's size cannot be checked, are skipped rather than charged, and before any audio is sent the longest the file can last (its real size at 32 kbps) is checked against your maximum cost per run; only the minutes Mistral reports are charged. Many shows also publish their own transcripts: see the transcripts field of each episode (free).
What is in the show profile?: The summary row of each show: title, author, owner name and e-mail as published in the feed, description, language, categories, explicit flag, show type, website, artwork, Apple id and genres, plus publishing statistics: episodes in the feed, first and latest episode dates, average duration, the median number of days between episodes and the episodes of the last 30 and 90 days.

Machine-readable

This page for agents

/actors/podcast-scraper.json — this API as JSON
/actors/podcast-scraper.md — this API as Markdown
/catalog.json — the whole catalogue