Command · Podcast Workflow
workflow.podcast
The one-click podcast / interview edit, on this computer (local, free): sync the cameras and microphones (multicam by audio), lay out the episode (the multicam picture, each microphone on its own dialogue track), transcribe, Enhance Speech, loudness to -16 LUFS, remove fillers and long pauses, switch cameras to whoever speaks, captions, chapters from the topics (and YouTube chapters), optionally shorts.
- v1
- Capability EditProject
- Tool
workflow_podcast
Request
{
"command": "workflow.podcast",
"params": {
"media": [
3,
4,
5,
6
]
}
}Parameters
anglesobject[]- Who is on which camera, for the switching (default: the analysis guesses).
angles[].angleinteger (1..64)Required- The angle: 1 = the first camera; the microphones come after the cameras.
angles[].role`closeUp` | `wide` | `unknown`- What it shows.
angles[].speakerstring- Whom a close-up shows: a speaker's name ("Mic 1": the first microphone's speaker) or key.
cameras(integer | string)[]- The camera recordings (the picture angles).
captionPresetstring- The caption preset (default Podcast).
contextobject- What earlier runs made (a previous result's "context", as it is).
context.multicaminteger (0..)- Multicam source id.
context.sequenceinteger (0..)- Episode sequence id.
context.transcribed(integer (0..))[]- What the run transcribed.
context.youtubeChaptersstring- The YouTube chapters text.
enhanceStrengthnumber (0.0..100.0)- Enhance Speech strength 0..100 (default 70).
fillersboolean- Remove filler words (default true).
lufsnumber (-36.0..-5.0)- Loudness target (default -16, podcasts).
maxChaptersinteger (2..100)- At most this many chapters (default 12).
media(integer | string)[]- Or all recordings: those with a picture are cameras, the rest microphones.
mics(integer | string)[]- The microphone recordings (audio only). Default: the cameras' own sound.
minChapterSecondsnumber (10.0..3600.0)- A chapter lasts at least this long (default 60 s).
namestring- The episode's name (default Podcast).
runboolean- Run the steps (default false: return the plan only).
shortsboolean- Also make vertical shorts of the best moments (default false).
shortsCountinteger (1..10)- How many shorts (default 3).
silencesnumber (0.0..30.0)- Shorten pauses longer than this many seconds (default 1; 0 keeps pauses).
skip(`multicam` | `sequence` | `transcribe` | `enhance` | `loudness` | `cleanup` | `autoswitch` | `captions` | `chapters` | `shorts`)[]- Steps to leave out.
steps(`multicam` | `sequence` | `transcribe` | `enhance` | `loudness` | `cleanup` | `autoswitch` | `captions` | `chapters` | `shorts`)[]- Run only these steps (with "run").
switchPreset`calm` | `interview` | `energetic`- The camera switching preset (default calm).
sync`audio` | `timecode` | `inPoints` | `outPoints` | `markers`- How the recordings are synced (default audio).
Discussion
The one-click podcast / interview edit, on this computer (local, free): sync the cameras and microphones (multicam by audio), lay out the episode (the multicam picture, each microphone on its own dialogue track), transcribe, Enhance Speech, loudness to -16 LUFS, remove fillers and long pauses, switch cameras to whoever speaks, captions, chapters from the topics (and YouTube chapters), optionally shorts. Without "run" it returns the plan (every step, on or off and why); with "run" it runs the enabled steps (or "steps"), as one undo step; "context" continues an earlier run.
Run it with wabcut-cli cmd <project> --json, from an MCP client as workflow_podcast, or from an extension. See Edit commands for requests, references and results.