Attesta
Home › Guides › Transcribe a podcast
How to transcribe a podcast
A transcript makes your podcast searchable, accessible, and easy to turn into show notes, clips, and social posts. The best approach depends on whether you are recording the episode yourself or working from an existing audio file. Here is how to do both, plus why speaker labels matter more for a podcast than almost any other kind of recording.
Transcribe an episode you record
If you are the host or a guest, the cleanest path is to transcribe as you record. With Attesta on iPhone:
- Tap record when you start. An audible tone plays, so your guest knows it is on.
- Give each speaker their own mic if you can, or keep everyone an equal distance from the phone — even levels make for cleaner diarization.
- Tap stop at the end.
- Get a speaker-labeled transcript plus a summary, key topics, and quotes — the raw material for your show notes.
Because it labels who said what, you can drop the transcript straight into show notes or pull standout quotes for clips.
What if the episode runs long?
Each recording captures up to 60 minutes before it stops automatically — plenty for a solo minisode, tight for a long-form interview episode. If you or a guest needs a break, pause instead of stopping: the clock stops with the recording and picks back up in the same file when you resume, so you are not left with a hard cut mid-sentence. If the conversation itself runs past 60 minutes, start a second recording where the first one leaves off — you end up with two transcripts back to back instead of one, but nothing is cut off silently.
Why speaker labels matter more for a podcast
A single-host monologue transcribes fine as a wall of text. A podcast with a host and one or more guests does not — without labels, a reader cannot tell who made which point. Compare a raw transcript of a two-person episode…
so today we are talking about pricing and I think the biggest mistake people make is charging by the seat instead of by usage that matches what I have seen too especially with teams that barely use the tool
…with the same audio, speaker-labeled:
Host: So today we are talking about pricing, and I think the biggest mistake people make is charging by the seat instead of by usage.
Guest: That matches what I have seen too — especially with teams that barely use the tool.
The labeled version is what you actually want for show notes, pull-quotes, and an accessible transcript page — you can attribute a quote to the right person without re-listening to check. It also holds up better with three or four voices on a panel episode, where an unlabeled transcript stops being usable almost immediately.
Get your guest's real name into the show notes
Diarization gives you accurate labels — Speaker 1, Speaker 2 — but that is not what belongs in public show notes, and it never changes: the transcript keeps those tags permanently, because they come from the audio itself, not a name (see how speaker labels work). What gets you from label to name is Refine, a Pro feature: tell Attesta who's who in plain language — "Speaker 2 is Jordan" — and that real name carries through automatically to the summary, key topics, and the quotes you pull for show notes. The in-app transcript still reads Speaker 1 and Speaker 2 either way, but the material you actually publish has your guest's name on it instead of a placeholder.
Transcribe an existing audio file
If the episode is already recorded and exported as a file, Attesta is not the right tool — it is a live recorder, not a file-upload service, so it cannot process an MP3 or WAV you already have. You want a dedicated file-based transcription tool instead. A few things worth checking before you pick one:
- Speaker separation. Podcasts almost always have more than one voice; a tool that only returns a flat transcript will cost you hours of manual labeling afterward.
- Pricing you can predict. Look for a transparent per-minute or per-file rate rather than a tiered plan that gates export or long files behind an upsell.
- Turnaround time. Some file-upload services queue longer files; check this against your publishing deadline, not just the marketing page.
- What happens to the audio afterward. If the file has anything sensitive in it (a guest's personal story, unreleased business details), check the provider's retention and training policy before uploading.
If you still have the conversation ahead of you — a future episode, or a re-record — recording it with Attesta skips the export/upload round-trip entirely: you get the speaker-labeled transcript and a structured summary the moment you stop recording.
Find a quote from three episodes ago
A few dozen episodes in, the hard part usually stops being the current recording and starts being the archive — a guest said something great, but you cannot remember which episode. Attesta's library search (Pro) checks the full text of every transcript you've recorded, not just the episode titles, so a half-remembered line turns up even when you cannot place which conversation it came from — the same search that works across your interview transcripts, if you record those too.
Turn the episode into show notes
Once the summary is ready, export it as a PDF straight from the recording — the AI summary, key topics, and quotes in one file, ready to build show notes from:
- Lead with the summary — a few lines on what the episode covers.
- List the key topics so listeners can jump around.
- Pull two or three quotes, attributed to the right host or guest by name if you've run Refine, for the episode description and social posts.
- Read the full transcript in-app for exact wording if you want to publish a complete transcript page — Apple Podcasts and Spotify both support them, and it helps you get found in search.
Record your episode and get a speaker-labeled transcript plus a summary and quotes — free to start, with an audible tone.
Download on theApp StoreFrequently asked questions
Can ChatGPT transcribe a podcast?
Not from audio on its own — it needs a speech-to-text step first and does not label speakers. A recording app like Attesta does audio-to-text in one pass, labels who said what, and writes a summary.
How do I transcribe my own podcast episode?
Record the conversation with an app and let it transcribe. Attesta gives you a speaker-labeled transcript plus a summary and topics to turn into show notes.
Can I upload an already-recorded episode to Attesta to transcribe it?
No — Attesta records live from your iPhone's microphone and does not accept uploaded audio files. For an episode you have already exported, use a file-based transcription tool instead. If you still have the conversation ahead of you, record it with Attesta and skip the export step.
How do I add a transcript to a podcast?
Generate it, clean up names, then upload to your host (Apple Podcasts and Spotify support transcripts) or paste it into your show notes. Attesta shows the full transcript in-app and exports the AI summary, key topics, and quotes as a PDF; for a full transcript page, read the in-app transcript for exact wording.
Is there a free way to transcribe a podcast?
Attesta has a free trial and pay-per-minute credits that never expire. Some web tools also offer limited free file transcription if you already have the audio.
Is there a recording time limit for podcast episodes?
Each recording captures up to 60 minutes before it stops automatically. Pause during a break or an interruption — the clock stops and resumes into the same file — and for a longer episode, start a second recording where the first one leaves off.
Can I rename Speaker 1 and Speaker 2 to my guest's real name?
Not in the transcript itself — the labels stay as Speaker 1, Speaker 2, and so on, since they come straight from diarization, not a name. Use Refine (Pro) to tell Attesta who's who — "Speaker 2 is Jordan" — and that real name carries through automatically to the summary, key topics, and quotes, the material you actually pull into show notes.
Can I find a quote from an old episode without remembering which one it was in?
Yes, on Pro — Attesta's library search checks the full text of every transcript you've recorded, not just the episode titles, so a half-remembered line turns up even when you can't recall which episode it was in.