---
title: "Podcast Transcript Generator with SRT and VTT | ToolSugar"
description: "Turn a podcast episode into a readable transcript with paragraph breaks, plus timed SRT and VTT files. The audio never leaves your browser."
canonical: https://toolsugar.com/tools/podcast-transcript
image: https://toolsugar.com/og.png
---

# Podcast Transcripts and Subtitles

ToolSugar turns a podcast episode into readable text. Whisper transcribes the MP3, WAV or M4A in your browser, then starts a new paragraph at any pause over 1.5 seconds or after 120 words. You also get SRT and VTT files for video versions. It does not label speakers, so you add names yourself.

Runs on your device

## How it works

1. Drop the episode file. Exported audio works, and so does a video recording of the session.
2. Run it on your device, which suits episodes that are not yet public. A long episode takes longer on a weak device.
3. Copy the transcript from the preview, or download the TXT, SRT and VTT files for your show notes.

## Key facts

- **Paragraphs**: A break at each pause over 1.5 seconds, with at most 120 words in a paragraph
- **Speakers**: Not identified. The transcript runs as one text with no speaker names
- **Long episodes**: The whole decoded recording is held in memory, about 230 MB per hour, so very long files need a capable device
- **Free runs**: Each transcription counts toward 5 free AI runs in any rolling 24 hours, no sign up

## Common uses

- Publishing a searchable transcript next to each episode for readers and search engines.
- Pulling quotes for a newsletter without replaying a full interview.
- Adding captions to the video cut of a show that is uploaded to YouTube.

## Questions

### Can it tell my co-host and guest apart?

No. The model does not separate speakers, so the text runs together with paragraph breaks at pauses. If you need names on each turn, add them by hand in your editor after copying the transcript.

### Is a one hour episode practical?

It can work, but it is heavy. Your browser has to hold the whole decoded recording in memory, about 230 MB per hour, and speed depends on your device. A weak laptop or phone may be slow or run out of memory on a long file. Splitting the episode into parts is safer.

### Will it remove filler words?

It writes what the small Whisper model hears and does not edit for style. It often drops some ums by itself, but you should not rely on that. Treat the transcript as a first draft that still needs a read through.

[Read the Podcast Transcripts and Subtitles guide](https://toolsugar.com/blog/how-to-add-subtitles-to-a-video)

## Related tools

- [Transcribe to Text](https://toolsugar.com/tools/transcribe-audio-to-text)
- [Auto Subtitles](https://toolsugar.com/tools/subtitle-generator)
- [Noise Remover](https://toolsugar.com/tools/noise-remover)
- [Audio to Video](https://toolsugar.com/tools/audio-to-video)