By Jason Snell
March 5, 2024 3:44 PM PT
Full transcripts arrive on Apple podcasts
As was foretold back in January, with the release of iOS 17.4 Apple’s Podcasts app now supports podcast transcripts. This is a pretty big breakthrough in terms of access to podcast content and accessibility of podcast to audiences who might not be able to listen.
The way Apple has implemented transcription is very clever. It’s all happening up in the cloud—the moment it detects that a new episode has arrived, Apple kicks that episode into its transcription queue and quickly generates a full transcript. (This is why, if you start listening the moment an episode drops, you won’t be offered a transcript—but very soon thereafter, it should appear.) Apple supports transcripts in English, Spanish, French, and German, which should cover 80 percent of overall listening in Apple Podcasts.
Apple’s not just running that podcast through a standard transcription engine like the one I use to generate transcripts on my Mac, but one that’s been built to detect some detailed information about how the podcast is structured.
That’s important, because many modern podcasts use something called Dynamic Ad Insertion to insert different ads depending on where you are, who you are, and when you downloaded the episode. A traditional transcript file won’t keep sync with a podcast if the time codes of the ads keep changing. Apple’s engine should be able to detect the beginning and end of those ads and adjust its transcript accodingly, inserting a filler animation (three slowly filling dots that will be familiar to users of lyrics in Apple Music) until the podcast content resumes, at which point the transcript should pick up right where it should.
Apple’s processing also detects content down to the word, so that (again, Apple Music style) it can highlight every word in the transcript as it’s spoken. It detects speaker changes and breaks paragraphs to improve readability, though it can’t identify the speakers. Episodes with chapter markers should see those reflected in the transcripts as subheads.
You can also select a paragraph from a transcript and share it (including a link back to the podcast), or even view the entire podcast transcript on its own without playing audio.
Podcasters who would prefer to use their own transcripts—I could see it happening in podcasts where there are some highly specific spellings and terms that they want to get exactly right—can do so by using the <podcast:transcript> field in their podcast RSS to point at a subtitles file in SRT or VTT format. Apple’s backend systems will pick that file up, run it through their own special processes, and supply it in the same interface.
The only thing that’s really missing is support for private podcast feeds, which is where most members-only versions of podcasts live these days. (Full disclosure: I produce several podcasts with members-only versions, and subscribe to several more!) I realize that there are some complicated technical isuses with members-only podcasts—technically each one is unique for each member, which is a real complicating factor—but between the file download URL and the URL of the transcript file, it should be doable for Apple to group all the members-only podcast episodes together. If it wants to transcribe those episodes itself, it’s more than welcome—but I’m also happy to provide my own transcript. I just don’t want my members missing out on this really great new feature.
If you appreciate articles like this one, support us by becoming a Six Colors subscriber. Subscribers get access to an exclusive podcast, members-only stories, and a special community.