Podcast Transcription: Episodes to Text, SRT & Monitor
Pricing
from $6.00 / 1,000 audio minute transcribeds
Podcast Transcription: Episodes to Text, SRT & Monitor
Transcribe podcast episodes to text: enter a podcast name, Apple Podcasts link or RSS feed and get the latest episodes as a transcript, SRT subtitles and timestamped segments. Monitor shows for new episodes. $6 per 1,000 audio minutes; failed episodes are free.
Pricing
from $6.00 / 1,000 audio minute transcribeds
Rating
0.0
(0)
Developer
Tenfold Fleet
Maintained by CommunityActor stats
0
Bookmarked
2
Total users
1
Monthly active users
4 hours ago
Last modified
Security
Limited permissionsShare
What does Podcast Transcription do?
Podcast Transcription turns podcast episodes into text. Type a podcast name, paste an Apple Podcasts link or an RSS feed URL, and get the full transcript of the newest episodes as plain text, an SRT subtitle file and timestamped segments, plus the episode title, publish date, duration, description and audio link.
Not affiliated with Apple or with any podcast publisher. Apple Podcasts is a trademark of Apple Inc.
Shows and RSS feeds, not files. This Actor finds episodes from a show name, Apple Podcasts link or RSS feed. To transcribe your own audio or video files from a URL, or from Google Drive, Dropbox or OneDrive links, use our Audio Transcription Actor. Both use the same speech engine and give the same transcript formats.
You don't need to find the MP3 yourself. The Actor looks the show up in the public Apple Podcasts directory, reads the podcast's own RSS feed, downloads the episode audio and runs AI speech to text (Whisper) on it, in about 100 languages. Long episodes are split into 10-minute parts behind the scenes and put back together, so a 2-hour interview comes back as one transcript with timestamps that run from 00:00 to the end.
Turn on monitor mode and schedule it: every run transcribes only the episodes that came out since the last run. It works as a podcast transcript API too: call it from your code, Make, Zapier or n8n and get JSON back.
Why use a podcast transcription tool?
- 📰 Track what is said on shows you follow: get each new episode of a news, industry or competitor podcast as text the morning it comes out, then search it or send it to an LLM for a summary.
- ✍️ Repurpose episodes: show notes, blog posts, newsletters, social clips and quotes from your own podcast or guests' appearances.
- 🎬 Subtitles: the SRT output drops straight into video editors and YouTube.
- 🤖 AI and RAG pipelines: timestamped segments are ready to chunk, embed and cite back to the exact minute.
- 🔎 Research: mentions of a brand, a person or a topic across many shows.
It runs on the Apify platform, so you also get scheduling, webhooks, an API and integrations with Make, Zapier, n8n, LangChain and LlamaIndex.
What data can it extract?
| Field | Example |
|---|---|
showTitle, showAuthor | Hard Fork, The New York Times |
episodeTitle, publishedAt | Why No One Can Look Away From Elizabeth Holmes, ISO date and time |
text | the full transcript as one string |
srt | the transcript as an SRT subtitle file |
segments | [{ "start": 600.01, "end": 612.11, "text": "..." }], times in seconds |
language, wordCount | en, 12808 |
episodeDurationSec, transcribedSec | length of the episode and of the part transcribed |
chargedMinutes, truncated | started minutes billed; true when only the first part was transcribed |
audioUrl, episodeUrl, episodeGuid | the episode's audio file, web page and feed id |
feedUrl, appleId, appleUrl, artworkUrl | where the show was found |
description, episodeNumber, season | from the feed |
error | why an episode or show could not be processed (those rows are free) |
How to transcribe a podcast episode
- Open the Actor and click Try for free.
- In Podcasts, add one show per line: a name like
Hard Fork, an Apple Podcasts show link, an Apple Podcasts episode link (with?i=in it) for one specific episode, an RSS feed URL, or a direct MP3 link. - Set Latest episodes per podcast (1 by default) and Max minutes per episode (120 by default).
- Pick the Output formats: plain text, SRT subtitles, timestamped segments, or all three.
- Click Start. A 1-hour episode usually takes 2 to 4 minutes.
- Download the results as JSON, CSV, Excel or HTML, or read them through the API.
To monitor podcasts for new episodes, turn on Only new episodes (monitor mode) and create a schedule (for example daily). Each run then transcribes only episodes it has not transcribed for you before; already-seen episodes are skipped for free.
How much does it cost to transcribe a podcast?
You pay per minute of audio transcribed: $0.006 per started minute, which is $6 per 1,000 audio minutes or about $0.36 for a 1-hour episode. The price does not depend on how many words are spoken.
- A show that can't be found, a feed that is down, an audio file that won't download, a failed transcription and an episode with no speech are free (you get a row with an
error). - Episodes skipped in monitor mode are free.
- Episode list only mode lists episodes with their metadata and audio link, without transcripts, for $0.002 per episode.
- Use Max minutes per episode and the run's maximum charge to cap spend. When the cap is hit mid-episode, you get (and pay for) the transcribed part, with a
notice.
Free Apify plan: up to 30 audio minutes per run, because every minute of speech to text costs us a fee. Any paid Apify plan removes the limit.
Input
{"shows": ["Hard Fork","https://podcasts.apple.com/us/podcast/the-daily/id1200361736","https://feeds.npr.org/500005/podcast.xml"],"maxEpisodesPerShow": 1,"onlyNewEpisodes": true,"maxMinutesPerEpisode": 120,"outputFormats": ["text", "srt", "segments"],"language": "auto"}
Output
One row per episode. Download the dataset as JSON, CSV, Excel, XML or HTML.
{"showTitle": "NPR News Now","showAuthor": "NPR","episodeTitle": "8pm ET, October 29 | The latest headlines from NPR","publishedAt": "2026-10-30T00:09:52.000Z","episodeDurationSec": 305,"transcribedSec": 304.7,"chargedMinutes": 6,"truncated": false,"language": "en","wordCount": 913,"text": "This is NPR News Now, the latest headlines every hour. Live from NPR News in Washington, I'm Rylan Barton. ...","srt": "1\n00:00:00,000 --> 00:00:06,300\nThis is NPR News Now, the latest headlines every hour.\n\n2\n...","segments": [{ "start": 0, "end": 6.3, "text": "This is NPR News Now, the latest headlines every hour." },{ "start": 6.82, "end": 10.04, "text": "Live from NPR News in Washington, I'm Rylan Barton." }],"episodeGuid": "85acd59f-8644-4105-a28a-0c428949146b","audioUrl": "https://.../episode.mp3","feedUrl": "https://feeds.npr.org/500005/podcast.xml","appleId": "121493675","error": null}
Tips
- Names are matched in the Apple Podcasts directory. If a name finds the wrong show, paste its Apple Podcasts link or RSS feed instead. Change Apple Podcasts country for shows listed only in another country's directory.
- One exact episode: open the episode in Apple Podcasts, copy its link (it contains
?i=) and paste it. - Long interviews: raise Max minutes per episode up to 300 for 3- to 5-hour episodes.
- Known language: setting
language(for examplees) helps on short or noisy episodes;autodetects it. - Monitor mode memory lives in your account's key-value store
podcast-transcription-seen. Delete that store to start fresh. - Video and short-form transcripts: for TikTok use TikTok Transcripts, for Instagram Reels use Instagram Reels Transcript, and for YouTube videos use YouTube Transcript Scraper.
FAQ
Which podcasts can it transcribe?
Any podcast with a public RSS feed, which is nearly every show in Apple Podcasts, Spotify, Pocket Casts and other apps. Shows that are exclusive to one app or behind a paid subscription have no public feed and come back as a free row with an error.
Can I paste a Spotify link?
No, Spotify pages don't expose the audio. Enter the show's name instead: most Spotify shows are also in the Apple Podcasts directory with a public feed.
How accurate are the transcripts?
The Whisper speech model is close to word for word on clear speech in common languages. Heavy crosstalk, music under the voice or poor audio cause mistakes, and timestamps are approximate. Speaker names are not identified. Ads read in the episode are transcribed like the rest of the audio.
Which languages are supported?
About 100, including English, Spanish, Portuguese, French, German, Italian, Dutch, Hindi, Japanese, Korean, Chinese, Arabic, Russian and Turkish. The language field gives the detected code.
Why is transcribedSec different from the length shown in the podcast app?
Many hosts insert ads into the audio file when it is downloaded, so the file can be a few minutes longer than the length in the feed. You pay for the minutes actually transcribed.
How does monitor mode know what is new?
It remembers every episode it transcribed for you (by the feed's episode id) in a named key-value store in your Apify account. Each scheduled run looks at the newest episodes of each show and transcribes only those it has not seen. Failed episodes are not remembered, so the next run tries them again.
Is it legal to transcribe podcasts?
The Actor only reads public RSS feeds and audio files that any podcast app can download. Podcasts are copyrighted: use transcripts for research, accessibility, search and analysis, and get permission before republishing someone else's show. You are responsible for complying with copyright law and the GDPR. If unsure, ask a lawyer.