How It Works

A Person Is Born

Every caller starts as a blank slate. At the top of each show the system writes a full roster of ten callers in a single pass — each with a name, age, hometown, and identity, plus the situation they're in, the reason they're calling tonight, and the first line out of their mouth. Each one also carries a secret want they never state outright, two or three specific details that could only belong to them, and an emotional register that sets how they hold themselves on the air.

The roster is built for contrast. A hard anti-collision rule keeps any two callers from sharing an obsession, a profession archetype, or a story theme — no two taxidermists, no two callers with ex-spouse drama, no two people chasing the same phenomenon. Each caller's hook appears exactly once in a roster, and the system rescans its own output to swap out anything that doubles up. Each caller is then matched to a voice that fits: a 60-year-old rancher outside Marathon doesn't sound like a 29-year-old grad student in Alpine.

Some callers become regulars. After a strong call, a first-timer can be promoted into a recurring character with their own file — a frozen set of canonical traits plus a running arc log that gets appended every time they call back. They remember past conversations, reference what they talked about before, and their stories move forward between episodes. Silas has spent his arc dragging The Wellspring out of New Mexico to a stretch of desert outside Terlingua, and the move is fracturing the community. Crispin, an MFA student at Sul Ross, is still living down the story he published about his ex. They're not reset between shows.

And some callers are drunk, high, or flat-out unhinged. They'll call with conspiracy theories about pigeons being government drones, existential crises about whether fish know they're wet, or to confess they accidentally set their kitchen on fire trying to make grilled cheese at 3 AM.

They Know Their World

Callers know real facts about where they live. The show broadcasts from Alpine, Texas, and the system carries accurate geographic knowledge of the Big Bend country of far West Texas — Alpine and Sul Ross State University, the Marfa Lights and the Chinati Foundation, the Gage Hotel in Marathon, the Terlingua ghost town and its chili cookoff, oilfield work up in Fort Stockton, and the Chisos Mountains and dark skies of Big Bend itself. Callers are held to real places only — they don't invent businesses or landmarks that aren't there. They know the current weather outside their window, what day of the week it is, how many hours it takes to drive anywhere out here, and what it's like living up against the border. They have strong opinions about their town and get nostalgic about how it used to be. The system also pulls in real-time news so callers can reference things that actually happened today.

They Have a Reason to Call

Some callers have a problem — a fight with a neighbor, a situation at work, something weighing on them at 2 AM. Others call to share a strange-but-true story, geek out about an obsession, or lay out a pattern only they can see. Nothing is drawn from a list: every caller's reason for calling is written fresh for that caller, tied to why they picked up the phone tonight specifically. Every caller has a purpose, not just a script.

The whole thing is tuned for comedy. Not "AI tries to be funny" comedy — more like the energy of late-night call-in radio meets stand-up meets the kind of confessions you only hear at 2 AM. Some calls are genuinely heartfelt. Some are absurd. Some start serious and go completely sideways. The system knows how to build a call for comedic timing — when to hold back a detail, when to escalate, when to let the awkward silence do the work. It's not random chaos; it's structured chaos.

The Conversation Is Real

Luke talks to each caller using push-to-talk, just like a real radio show. His voice is transcribed in real time, sent to an AI that responds in character, and then converted to speech using a voice engine — all in a few seconds. The AI doesn't just answer questions; it reacts, gets emotional, goes on tangents, and remembers what was said earlier in the show.

Callers don't just exist in isolation — the show tracks what's been discussed and matches callers thematically. If someone just called about a messy divorce, the next caller who references marriage didn't pick that topic randomly. The system scores every previous caller against the current one for topic overlap and decides whether the new caller should reference them, disagree with them, or build on what they said. The strength of the match sets how likely that reaction is: a strong thematic hit fires most of the time, a weak one only occasionally, so callbacks land when they're earned instead of on every single call.

And when a call has run its course, Luke can hit "Wrap It Up" — a signal that tells the caller to wind things down gracefully. Instead of an abrupt hang-up, the caller gets the hint and starts wrapping up their thought, says their goodbyes, and exits naturally. Just like a real radio host giving the "time's up" hand signal through the glass.

Real Callers Call In Too

When you dial 208-439-LUKE, your call goes into a live queue. Luke sees you waiting and can take your call right from the control room. Your voice streams in real time — no pre-recording, no delay. You're live on the show, talking to Luke, and the AI callers might even react to what you said. And if Luke isn't live, you can leave a voicemail — it gets transcribed and may get played on a future episode.

Listener Emails

Listeners can send emails to [email protected] and have them read on the show. A background poller checks for new messages every 30 seconds — they show up in the control room as soon as they arrive. Luke can read them himself on the mic, or hit a button to have an AI voice read them aloud on the caller channel. It's like a call-in show meets a letters segment — listeners who can't call in can still be part of the conversation.

Devon the Intern

Every show needs someone to yell at. Devon is the show's intern — a 23-year-old NMSU grad who's way too eager, occasionally useful, and frequently wrong. He's not a caller; he's a permanent fixture of the show. When Luke needs a fact checked, a topic researched, or someone to blame for a technical issue, Devon's there.

Devon has real tools. He can search the web, pull up news headlines, look things up on Wikipedia, and read articles — all live during the show. When a caller claims that octopuses have three hearts, Devon's already looking it up. Sometimes he interjects on his own when he thinks he has something useful to add. Sometimes he's right. Sometimes Luke tells him to shut up. He monitors conversations in the background and pipes up with suggestions that the host can play or dismiss. He's the kind of intern who tries really hard and occasionally nails it.

The Control Room

The entire show runs through a custom-built control panel. Luke manages callers, plays music and sound effects, runs ads and station idents, monitors the call queue, and controls everything from one screen. Audio is routed across eight independent channels simultaneously — host mic, AI caller voices, Devon, live phone audio, music, sound effects, ads, and station idents all on separate tracks. The website shows a live on-air indicator so listeners know when to call in.

Multi-Stem Recording

During every show, the system records seven separate audio stems simultaneously: host microphone, AI caller voices, Devon, music, sound effects, ads, and station idents. Each stem is captured as an independent WAV file with sample-accurate alignment. This gives full control over the final mix — like having a recording studio's multitrack session, not just a flat recording.

Dialog Editing in REAPER

Before the automated pipeline runs, the raw stems are loaded into REAPER for dialog editing. A custom Lua script analyzes voice tracks to detect silence gaps — the dead air between caller responses, TTS latency pauses, and gaps where Luke is reading the control room. The script strips these silences and ripple-edits all tracks in sync so ads, idents, and music shift with the dialog cuts. Protected regions marked as ads or idents are preserved — the script knows not to remove silence during an ad break even if the voice tracks are quiet. This tightens a raw two-hour session into a focused episode without cutting any content.

Post-Production Pipeline

Once the show ends, a 15-step automated pipeline processes the raw stems into a broadcast-ready episode. Ads and sound effects are hard-limited to prevent clipping. The host mic gets a high-pass filter, de-essing, and breath reduction. Voice tracks are compressed — the host gets aggressive spoken-word compression for consistent levels, callers get telephone EQ to sound like real phone calls. All stems are level-matched, music is ducked under dialog and muted during ads, then everything is mixed to stereo with panning and width. A bus compressor glues the final mix together before silence trimming, fades, and EBU R128 loudness normalization.

Automated Publishing

A single command takes a finished episode and handles everything: the audio is transcribed using MLX Whisper running on Apple Silicon GPU to generate full-text transcripts, then an LLM analyzes the transcript to write the episode title, description, and chapter markers with timestamps. The episode is uploaded to the podcast server and directly to YouTube with chapters baked into the description. Chapters and transcripts are attached to the RSS metadata, all media is synced to a global CDN, and social posts are pushed to eight platforms — all from one command.

Automated Social Clips

No manual editing, no scheduling tools. After each episode, an LLM reads the full transcript and picks the best moments — funny exchanges, wild confessions, heated debates. Each clip is automatically extracted, transcribed with word-level timestamps, then polished by a second LLM pass that fixes punctuation, capitalization, and misheard words while preserving timing. The clips are rendered as vertical video with speaker-labeled captions and the show's branding. A third LLM writes platform-specific descriptions and hashtags. Then clips are uploaded directly to YouTube Shorts and Bluesky via their APIs, and pushed to Instagram Reels, Facebook Reels, Mastodon, Nostr, LinkedIn, Threads, and TikTok — nine platforms, zero manual work.

Global Distribution

Episodes are served through a CDN edge network for fast, reliable playback worldwide. The RSS feed is automatically updated and picked up by Spotify, Apple Podcasts, YouTube, and every other podcast app. The website pulls the live feed to show episodes with embedded playback, full transcripts, and chapter navigation — all served through Cloudflare with edge caching. From recording to available on every platform, the whole pipeline is automated end-to-end.