PodcastsToText
freakonomics radio transcript

Freakonomics Radio transcripts

A produced, edited, multi-voice show, which changes what a transcript is good for.

Host
Stephen J. Dubner
Episode length
Usually around forty to sixty minutes.
Format
Heavily produced narrative journalism: a narrator, interview tape from several people, music and sound design.

Why people transcribe this show

Freakonomics Radio is the odd one out here. It is not a conversation recorded in one take; it is edited narrative journalism, with a host narrating over interview clips from several people, cut together tightly. That makes it much shorter than the others and much denser, an hour of this show contains more distinct claims than three hours of open conversation.

For research and citation that density is the point. If you are following an argument, or tracking down which economist said which thing, the text is far more useful than the audio, because the speakers change every thirty seconds.

How to do it

Three steps, and the first one is the only one with a decision in it.

  1. Get a link to the episode. Copy the episode link from Spotify or Apple Podcasts: the share link for the individual episode, not the show. A direct link to an audio file works too, and so does uploading the file yourself if you already have it.
  2. Paste it and pick a format. Plain text for reading and notes, SRT or VTT if the words are going on video, JSON if something else is going to read it. Speaker labels are worth leaving on for anything with more than one voice.
  3. Read, correct, export. Fix the proper nouns (see below for where they go wrong on this show), then export. The transcript stays in your library, so you only pay the minutes once.

What's awkward about Freakonomics Radio

The parts worth knowing before you spend the minutes.

Many speakers, short turns

Narration cut with interview tape means a lot of speaker changes, often only a sentence or two each. Speaker separation handles the switch itself well, but it cannot know who anyone is, labels come out as generic speakers, and you name them yourself in the editor.

Music and sound design under speech

Produced shows put music beds and effects under dialogue. Modern transcription handles this far better than it used to, but a line delivered over a loud bed is the most likely place to find an error. They are usually obvious in context.

Shorter episodes are cheap to transcribe

At forty to sixty minutes these cost a fraction of a long-form episode, and they fit inside a free monthly allowance in a way a three-hour conversation never will. If you are trying this out, a produced show is the cheapest place to start.

FAQ

Questions about Freakonomics Radio

It will tell you that the speaker changed, and keep them consistent, but it cannot know their names, it has never heard them introduced by voice. You rename the generic labels in the editor, once per person, and they apply throughout.

Other shows

Transcribe a Freakonomics Radio episode

Paste an episode link and read it in about a minute. Thirty minutes a month are free, with no card.

PodcastsToText is not affiliated with, endorsed by, or sponsored by Freakonomics Radio or Stephen J. Dubner. We do not host or distribute transcripts of this show. This page explains how to generate your own transcript from an episode you already have access to.