Search "free podcast transcription" in 2026 and you get pages of tools that promise free transcription, then hit you with a watermark, a 5-minute limit, or an upgrade wall the moment you try to export the text. A handful of them are genuinely free and genuinely good. A few more are free-ish if you know what you're doing. Everything else is a paid product wearing a free sticker.

This guide walks through what actually works, ranked by how much effort they cost and how good the output is. No referral links, no "free trial" padding. Just the real options.

The short version

If you want a transcript of a single podcast episode right now for free:

  • If the show is on Apple Podcasts → Use Apple's built-in auto-transcript. You already have it and didn't realize.
  • If the show is on YouTube → Use YouTube's auto-captions and copy the transcript from the description panel.
  • If neither → Download the MP3 and run it through OpenAI Whisper, which is free and open source, if you're comfortable with the command line.

That covers 90% of real needs. Everything below is the longer version for the 10%.

Method 1: Apple Podcasts auto-transcripts (genuinely free, already done for you)

Apple started auto-generating transcripts for most shows in its catalog in 2024. By 2026 the coverage is nearly universal for any show distributed through Apple Podcasts. If you listen on iPhone or Mac, you almost certainly have the transcript one tap away and never noticed.

How to use it: Open the episode in Apple Podcasts. Tap the quote-bubble icon (the little speech-bubble glyph near the playback controls). You get a scrolling, time-synced transcript that highlights the current line as the episode plays. You can tap any line to jump to that point. You can search within the transcript.

What you can't do: Export the full text cleanly. Apple really wants you to read the transcript inside their player, not copy it out. You can select and copy sections of text, but there's no "export full transcript" button. For casual reading, that's fine. For research or quoting, it's frustrating.

Cost: Free. You already paid for it (it's part of Apple Podcasts).

When it fails: A small number of shows opt out. Independent shows hosted outside Apple's system sometimes don't have auto-transcripts yet. Non-English shows have much worse coverage.

Method 2: Spotify podcast transcripts (free, coverage is growing)

Spotify launched podcast transcripts in 2023 and coverage has grown steadily but unevenly. Big shows have them, independent shows often don't. Same limitation as Apple — the transcript lives inside the Spotify app and exporting is awkward.

How to use it: Open the episode in the Spotify mobile app, start playback, and the transcript appears under the player controls. If no transcript is available, Spotify doesn't show the button at all, so absence means no transcript rather than a hidden feature.

Cost: Free with a Spotify account.

When it fails: Coverage is patchy for shows outside Spotify's own network (Joe Rogan, Bill Simmons, etc.). Non-English is worse. Smaller independent podcasts are often missing entirely.

Method 3: YouTube auto-captions (free, works for any show that's also on YouTube)

A surprising number of podcasts now publish their episodes on YouTube as well — usually as static-image videos with the audio waveform. YouTube auto-generates captions for any uploaded video, and those captions are free, exportable, and searchable.

How to use it:

  1. Find the episode on YouTube. Most major podcasts have a YouTube channel; search "[show name] [episode name/number]" and you'll find it.
  2. Open the video. Click the three-dot menu under the video and select "Show transcript."
  3. The transcript appears in a side panel. You can copy the whole thing — click inside the panel, Ctrl+A, copy, paste into a document.

Cost: Free.

Quality: Usually decent for English shows with clear audio. Struggles with overlapping speakers, strong accents, and technical jargon. No speaker attribution — you get the words but not who said them, which is a problem for interviews.

When it fails: The show isn't on YouTube. The episode is on YouTube but uploaded by a fan (fan uploads sometimes have the wrong audio or are incomplete). The show is on YouTube but auto-captions are disabled by the uploader.

Method 4: OpenAI Whisper (genuinely free, best quality, more effort)

Whisper is OpenAI's open-source speech-to-text model. It runs on your computer, no API calls, no usage limits, no watermark. Quality is better than any of the web-based free options. The catch is that you need to be comfortable with the command line and willing to wait while it processes.

How to use it:

  1. Download the episode as an MP3. Most podcast apps let you download episodes directly; you can also grab the MP3 URL from the show's RSS feed.
  2. Install Whisper. The simplest path is pip install openai-whisper on a machine with Python. On a Mac, you can also install via Homebrew.
  3. Run whisper episode.mp3 --model medium and wait. A one-hour episode takes roughly 10–30 minutes to transcribe on a modern laptop, depending on your hardware and which model size you pick.
  4. Whisper outputs the transcript in multiple formats — plain text, SRT subtitles, VTT, and JSON with timestamps. Pick whichever fits your use case.

Cost: Free. The model weights are open source, there's no API, and there's no account required.

Quality: Very good. For English with clear audio, Whisper produces transcripts that rival paid human-edited services. It handles multiple speakers reasonably well, though without speaker labels by default — you get what was said, not who said it. For speaker attribution you need a separate tool to "diarize" the output.

When it fails: Your computer doesn't have enough RAM (the larger models want 10+ GB). You don't have command-line comfort. You need the transcript in the next 5 minutes and don't have time to wait for processing.

Method 5: Free tiers of paid tools (free-ish, with limits)

Most paid transcription tools offer a free tier with meaningful limits. Worth mentioning because they're what most people land on when they search "free podcast transcription" and they're usually fine for one-off needs — just don't expect to transcribe a full backlog on the free plan.

Otter.ai — Free tier: 300 minutes of transcription per month, 30 minutes per audio file. Real-time transcription if you stream the audio into the app. Speaker separation and decent accuracy.

Descript — Free tier: 1 hour of transcription per month. Quality is high and the editing interface is excellent if you want to work with the transcript afterward.

Trint — Free tier: very limited, mostly a trial rather than a real free plan.

The general rule is that free tiers on paid products work if you transcribe 1–2 episodes a month and are willing to manage an account per tool. They don't work for ongoing use — the paid wall hits fast.

What to avoid

Several categories of "free podcast transcription" tools are not worth your time.

"Free online podcast transcription" websites that require upload. Most of these cap at 5 minutes, watermark the output, or secretly route your audio to a cheap transcription service that then tries to sell you a subscription. The economics don't work for actually free — transcription costs compute — so if the landing page promises "unlimited free transcription," be skeptical.

Browser extensions that claim to transcribe any audio. Most are privacy disasters (they upload your audio to someone else's server) or they use YouTube's captions under the hood and charge for a feature you could use directly for free.

AI chatbots you paste the MP3 into. ChatGPT, Claude, and similar tools can transcribe audio but usually have tight file-size limits and inconsistent quality. Good for a 60-second clip, bad for a 90-minute episode.

When you don't actually want a transcript

If you've read this far, here's the thing worth saying: most people who search for "free podcast transcription" don't actually want a transcript. They want to remember what was said, find a specific part, or get the information out of the episode without having to listen to the whole thing. A transcript is one way to do that, but for an hour-long episode, it's an 8,000-word wall of text you're never going to read in full.

If your real goal is extracting the information from a podcast — not producing a text document — a transcript is the wrong tool. A text-summary is closer, but it flattens the tonal information that made the podcast worth listening to in the first place. The best format for the information in an audio source is a shorter version of the audio itself. A briefing.

This is what TrimCast is built for. You paste a podcast URL and get back a condensed multi-voice audio version of the episode that covers the same ground in 10 to 20 minutes, with speaker attribution and key quotes intact. No transcription step, no text wall, no lost tone. If you still want the underlying transcript alongside the briefing, you get it — but most people stop needing the transcript once they have the briefing.

Pick the method that fits

  • One episode, right now, free, zero setup → Apple Podcasts or YouTube auto-captions
  • Lots of episodes, best quality, free forever, willing to set up tools → Whisper
  • One episode every month or two, willing to deal with an account → Otter or Descript free tier
  • You want the information, not the text → Try TrimCast. Paste the URL, get a listenable briefing in minutes. Actually free to try.

Your AI podcast assistant. Listen smarter, not longer.