How to edit a video podcast: from raw tracks to YouTube, Spotify and clips

To edit a video podcast, sync the cameras and audio, cut between speakers, tighten the conversation by editing the transcript (false starts, crosstalk, long tangents, stalling fillers), fix the sound, add chapters, and export a 16:9 episode for YouTube and Spotify plus vertical clips for social. Most of the time goes into the conversation edit, which is why transcript-based editors such as Descript and Riverside have become the default for podcasts.

This guide covers the whole episode. For short social clips from it, see how to make podcast clips.

Before you edit: record for the edit

Editing gets easier the more you separate at recording time:

• Separate audio tracks per speaker. You can fix one person’s levels or remove a cough without touching the others.
• Separate camera per speaker, or at least one wide and one close shot.
• A clap or a sync point at the start if cameras and audio record separately.
• Local recording for remote guests. Platforms such as Riverside record each participant locally and upload the files, so a shaky internet connection does not end up in the episode (Riverside).

The edit, step by step

1. Sync and assemble

Bring in all camera angles and audio tracks and line them up on the sync point. Most desktop editors can sync by audio waveform automatically. Then build a multicam sequence, so you can switch between angles as you watch.

2. The conversation edit

This is the main work, and it is about words, not pictures:

• Remove the warm-up and the wrap-up that nobody needs: “are we recording?”, technical checks, the long goodbye.
• Cut false starts and restarts. “So the, the thing is, what I mean is” becomes “The thing is”.
• Cut tangents that do not pay off. A podcast can afford some, but not ten minutes on parking.
• Clean up crosstalk where both people talk at once and neither can be understood.
• Remove fillers that stall, keep the ones that sound like conversation. A podcast stripped of every “um” sounds scripted (filler words are not the enemy).
• Tighten long pauses, but leave the natural rhythm of conversation.

In a transcript editor, this step is reading: you strike text and the audio and video follow. Descript detects filler words and lets you remove them per word or all at once, optionally leaving a gap to keep the timing (Descript Help). Riverside’s editor removes filler words with a smart, cut or mute mode (Riverside Support). The details are in how to remove filler words from a video.

3. Camera cuts

With the conversation set, choose the angle for each moment: the speaker when someone makes a point, the listener for a reaction, the wide shot to reset. Several podcast tools can switch automatically to whoever is speaking; check the result, because automatic switching misses reactions.

4. Sound

• Even out levels between speakers.
• Reduce background noise and room echo, gently.
• Set the loudness for the platform. YouTube and Spotify are commonly cited at around −14 LUFS and Apple Podcasts at around −16 LUFS (Descript); platforms turn louder files down, so a consistent target matters more than a loud one.
• Keep music to the intro, outro and transitions.

5. Chapters, titles and graphics

Add chapters with timestamps for YouTube, a lower third with the guest’s name, and graphics only where they explain something.

6. Export

• YouTube: 16:9, at the resolution you recorded.
• Spotify: Spotify for Creators accepts video episodes; guides to its requirements list MP4 with H.264 video and AAC audio, with 1080p recommended (The Podcast Consultant). Check Spotify’s own help pages before you finalise settings.
• Audio-only feeds: export the audio separately at your platform’s loudness.
• Clips: vertical 9:16, 30–90 seconds, with captions.

Which software for a video podcast

Software	Platform	Strength	Price as of October 2026
Descript	Mac, Windows, browser	The most mature transcript editor; filler words, multitrack, audio cleanup	Free plan with watermark; paid from $16/month billed annually (Descript)
Riverside	Browser	Remote recording with local tracks, text-based editor, Magic Clips	Free plan with watermark; Pro $29/month or $24/month billed annually (Riverside)
Adobe Premiere	Desktop	Full multicam control, Text-Based Editing for speech	Subscription
DaVinci Resolve	Desktop	Free version with multicam; Fairlight for audio	Free; Studio is a one-time purchase

How the two podcast specialists compare: Descript vs Riverside.

Where slicecut fits, and where it does not

slicecut is a phone editor for short talking videos. A source video can be up to 15 minutes and a finished project up to 20 minutes. That rules it out for a 60-minute episode, and it has no multicam or chapters. For full episodes, use one of the desktop tools above.

It does fit two podcast jobs:

• Clips. Import the segment around a strong moment, strike everything around it in the transcript, tighten the pauses, export at source quality. See how to make podcast clips.
• Short video episodes and solo shows under 20 minutes, recorded on a phone: cut false starts and repeats by text, set the pauses with one dial, export.

Planning your time

Times vary with the show, but as a rough plan for a one-hour conversation in a transcript editor: about an hour to read and cut the conversation, half an hour for camera choices, half an hour for sound and export, and another hour for three to five clips. The conversation edit is the largest part, and it is the part where reading beats listening by the widest margin.