# Podcast transcript with speakers

URL: https://neattrim.com/tools/podcast-transcript

Upload an episode and get the full text, divided by speaker, in a format you can publish or edit.

## A transcript you can publish, search and quote

Most podcast listeners never read a transcript, but the people who do are the ones who matter for reach. A transcript makes an episode searchable, gives search engines the words, lets deaf and hard-of-hearing listeners follow, and makes it simple for a journalist to quote you correctly. Speaker labels are what turn a wall of text into something readable, because an interview with no names on the lines is hard to follow.

This preset does one job. It transcribes the episode, tells the speakers apart and stops. It does not clean the audio, write show notes or add chapters, so it runs sooner and uses less. Use it when the episode audio is already finished.

Speakers are labelled in the order they first appear. Detection separates voices by sound, so two guests with similar voices, or a guest who speaks only a few lines, can be merged or split wrongly. Skim the first few exchanges and fix any swapped labels in the export.

The podcast JSON output is the format that podcast apps read for timed transcripts, and DOCX is useful when an editor wants to tidy the text before it goes live.

## Steps

1. **Add the episode.** Drop in an MP3, M4A, WAV or video file of the episode. A single mixed file is fine.
2. **Keep speaker labels on.** This page switches speaker detection on and runs only the transcript step. No clean-up, notes or chapters are made.
3. **Choose the language.** Leave it on auto, or set the spoken language if the episode is not in English.
4. **Export for publishing.** Download TXT for a web page, DOCX for an editor, or the podcast JSON that podcast apps use for transcripts.

## Questions

**How does the tool know who is speaking?**

It compares the sound of each voice and groups similar segments. It does not know names, so the labels are Speaker 1, Speaker 2 and so on.

**Can the speaker labels use real names?**

Labels are generic, so replace them with real names using find and replace in your editor.

**What if a guest joins by phone and sounds very different?**

A changed sound path can split one person into two speakers. Check the labels near the guest's first and last lines.

**Does the transcript include filler words?**

It keeps what was said, so ums and repeated words are in the text. Edit them out in the document if you want a tidier read.

**Which file is the podcast JSON for?**

It is a timed transcript file that podcast apps can display while the episode plays. Link it from your feed if your host supports transcript tags.
