What Pod Monster reads
The long version of what the landing page introduces: where each one comes from in a feed, and how much of it a player is expected to show.
The chapters the creator intended
Chapters are older than podcasting itself. Apple built them for QuickTime in the mid-90s, then reused the format for audiobooks (M4B) and podcasts (M4A/AAC) when iTunes 4.9 shipped in June 2005. That December, ID3 gained chapters, bringing them to MP3 — though Apple didn't display those until around 2018.
Everything since has moved chapters out of the file and into the feed. Podlove Simple Chapters arrived in 2009 as an RSS extension, and stayed common in European shows; Podcasting 2.0 proposed podcast:chapters JSON in 2020, which Apple began ingesting with iOS 26.2 in December 2025.
Every app supports a different slice of that history, and some now generate chapters over the top of the ones a show wrote. Pod Monster reads the creator's chapters, keeps art and titles intact, and caches them with a download so the list works offline. Any autogeneration of chapters that's done in the future will only be for episodes with none.
| Chapter format | Where it lives | Since | Pod Monster | Apple |
|---|---|---|---|---|
| MP4 / M4B chapters | Inside the AAC or MP4 file | 2004 | no | yes |
| MP3 / ID3 chapters | Inside the MP3 file | 2005 | yes | yes |
| Vorbis chapters | Inside the Ogg, Opus or FLAC file | no single date | no | no |
| Podlove Simple Chapters | In the RSS feed | 2009 | yes | no |
| podcast:chapters (JSON) | In the RSS feed | 2020 | yes | yes |
| Timestamps in the notes | In the episode description | no single date | yes | yes |
| Auto-generated | Made by the app | 2025 | no | yes |
Apple added ID3 support around 2018 and the rest with iOS 26.2 in December 2025. Vorbis chapters use a comment convention with no single dated spec; support is on our list. Dates are when each format appeared, not when apps adopted it.
Follow the words as they play
Transcripts have their roots in broadcast television closed captioning, which the US and the UK developed throughout the 1970s and 80s. But the oldest timed format that podcasts still use today was created in 2000, when a developer who went by Brain released SubRip. The program read the subtitles off of a video or a DVD and saved them as a plain text SRT file: a number, a start and end time, and the words.
In July 2010, Ian Hickson drafted WebSRT to bring SRT to the web, and he renamed it WebVTT that December. Podcasting 2.0 gave feeds podcast:transcript in 2020, so a show could link a transcript in VTT, SRT, HTML, plain text, or a new JSON format. Apple added transcripts to the Podcasts app with iOS 17.4 in March 2024. It generates its own, and it reads a show's file only in VTT or SRT, and only after the creator opts in.
Pod Monster reads all five, with their time codes and speaker names. Tap a line to jump to it, or read the episode without playing it. When a show inserts ads into a downloaded episode, Smart transcripts listens to the audio on your device and moves the times to match.
We don't generate transcripts today, and if we ever do, we'll only use them for shows that publish none, never in place of a show's own.
| Transcript format | Time codes | Speaker names |
|---|---|---|
| WebVTT text/vtt | A start and an end on every cue | A <v> voice tag, a [SPEAKER_00] label, or a Name: prefix |
| SubRip (SRT) application/x-subrip | A start and an end on every cue | A [SPEAKER_00] label or a Name: prefix |
| JSON application/json | startTime and endTime, in seconds | The speaker field of each segment |
| HTML text/html | Stamps in the text, or a <time> element | A <cite>Name:</cite> label, or Name [00:01:14]: |
| Plain text text/plain | Stamps in the text, such as [00:01:14] | Name [00:01:14]:, or a [Speaker 1] label |
When a feed lists more than one transcript, Pod Monster tries VTT first, then SRT, then JSON, then HTML and plain text, and uses the first one that reads.
Time codes
VTT, SRT, and JSON give every line a start and an end, and Pod Monster uses them as written. HTML and plain text only stamp where a paragraph starts, so each line runs until the next stamp. When two speakers share one stamp, Pod Monster splits the time by how much each one says. A file with fewer than two stamps has no times of its own.
Speaker names
A label the format marks as a name, such as a voice tag, the JSON speaker field, or an HTML cite, holds until the next one. A plain Name: at the start of a line names that line only, unless the file uses two or more of them. A bracket label must say "speaker", so [MUSIC] stays in the text.