Free tool
Podcast Transcription and Show Notes
Drop in a finished episode and get its transcript with the host and guests on every line, then build show notes from it on this page: chapters that open at a speaker turn, a title for each, and the share of speaking time per voice. The first 60 minutes are free with no account.
Podcast to transcript: an example
Input: the first minutes of a 48-minute episode with a host and two guests (episode-31.mp3).
Output:
| Time | Speaker | Text |
|---|---|---|
| 00:06 | Maya Chen | Welcome back. Today I'm joined by Tom Becker and Priya Nair. |
| 00:13 | Tom Becker | Thanks for having me, Maya. |
| 00:16 | Priya Nair | Happy to be here. |
| 00:21 | Maya Chen | Tom, you started the studio in 2019. What was the first thing you built? |
| 00:29 | Tom Becker | A small tool for editing audio, honestly. It grew from there. |
All three are named from the introduction. A guest who is never named would show as a letter.
Show notes with chapters, from the timestamps
Pick how long a chapter should run at least. Each one opens at the first speaker turn after that point, never mid-sentence, and its working title is the first words said there.
Showing an example 45-minute episode. When your transcript is ready above, this switches to it.
- 00:00
- 10:40
- 21:50
- 33:10
- 44:10
With Maya Chen, Tom Becker, Priya Nair In this episode 00:00 Welcome back. Today I’m joined by Tom Becker… 10:40 Priya, how did you two end up working… 21:50 Let’s talk about pricing. When did you raise… 33:10 Last question for both of you 44:10 Where can people find you? Thanks to both…
Naming co-hosts and guests
A voice gets a name when someone says it on the recording. Guests are usually introduced, so they come out named; co-hosts who never address each other by name come out as letters. Rename a letter once — the show notes above, the copy and every download use the new name.
The percentages over the chapters are each voice's share of the speaking time. On an interview episode the guest usually holds most of it; if the host has more, that is worth knowing before the next one.
Which export to upload
| What you have | Upload it? |
|---|---|
| The final mix you published, usually an MP3 | Yes — the timestamps then match the episode listeners hear, so chapter times are right. |
| The raw recording before editing | Only for your own notes: every cut you make later shifts the times after it. |
| One track per person, recorded separately | Mix them to one file first. A single track holds one voice, so the whole transcript would be one speaker. |
| A video podcast as .mp4 | Export the audio only — a long video soon passes the 350 MB limit, and the picture adds nothing to the text. |
In Audacity that is File → Export Audio; in GarageBand, Share → Export Song to Disk. Downloading the episode from your hosting dashboard gives the same final mix.
The upload limit, the 60 free minutes and when the audio and the text are deleted are the same as on the audio to text converter, which sets them out in full.
From episode file to show notes
- Export the final mixThe MP3 you published, or the episode downloaded from your hosting dashboard. Separate tracks per person go in mixed to one file.
- Fix the namesGuests are named from the introduction. Give a name to any co-host who came out as a letter — the show notes pick it up.
- Build the show notesChoose how long a chapter runs, rewrite each working title, and copy the notes with their timestamps under the episode.
Frequently Asked Questions
By time and by turns. You pick the shortest a chapter may run — 5, 10 or 15 minutes — and each new chapter starts at the first change of speaker after that. The title is the first words said there; rewrite it into the topic before you publish. The first chapter always starts at 00:00.
Yes. Each voice gets its own label, and a guest is named as soon as someone says the name aloud. Voices that sound alike, or people talking over each other, can end up under one label, so check the names against the introduction.
A mix of all the tracks in one file. Each separate track holds one voice, so a transcript of it has one speaker and none of the questions; the mix has everyone in order.
Yes. Download SRT or VTT and upload it to your video host, or use the TXT or Word file for a page on your site. Read the text once before publishing: unusual names and brand words are the likeliest errors.
No, the page takes a file. Download the episode as MP3 from your hosting dashboard, or from the audio link in your feed, and drop it here.
Music without words leaves no lines. Sung lyrics and ad reads are speech, so they are transcribed like the rest; delete those lines from the text if you do not want them in the show notes.
