A quick note before we start: Jordan Ellis and the podcast Bootstrapped Founders Weekly are an illustrative composite, not a single verified interview. We built this example from patterns we see again and again across the solo and small-team podcasters who use PlayHT's voice cloning to rebuild their production process. If it sounds familiar, that's the point.

The old workflow: a lot of hours for one episode

Jordan records two 40-minute interview episodes a week for Bootstrapped Founders Weekly, a show about early-stage SaaS founders. Before touching PlayHT, a single episode looked like this:

  • Record the interview (45–60 minutes, including false starts)
  • Edit the raw audio, cut dead air, level the levels (60–90 minutes)
  • Re-record a fresh intro and outro every single episode, often multiple takes because of flubbed lines, background noise, or a scratchy voice on a Monday morning (30–45 minutes)
  • Write show notes and a summary for the episode page (45 minutes)
  • Cut two or three short audiogram clips for Instagram and TikTok, then either skip the voiceover or record one separately (60 minutes)
  • Occasionally record a short bonus segment for paying subscribers, when there was time (30–45 minutes)

Add it up and Jordan was spending close to 14–15 hours a week on production for just two episodes, on top of the actual interviews. The intro and outro re-recording was the most demoralizing part of the routine: the same 90 seconds of scripted copy, recorded from scratch, week after week, because a host's voice changes depending on the time of day, how tired they are, or whether they're fighting a cold.

The turning point: cloning the intro voice

Jordan's first experiment with PlayHT wasn't ambitious. It was a Friday afternoon test: upload a few clean minutes of solo narration, create a cloned voice, and see whether a generated intro was good enough to actually ship. It was. The cadence, the pauses between sentences, even the slight rasp Jordan gets by the end of a long recording day — the clone carried it over convincingly enough that a returning listener wouldn't have noticed the switch. From there, the workflow expanded piece by piece, one re-recording headache at a time.

The skepticism didn't disappear overnight. Jordan ran the first few AI-generated intros past a small group of regular listeners without telling them anything had changed, and asked afterward if anything sounded off. Nobody flagged it. That informal test is what turned a curiosity into a permanent part of the workflow.

The new workflow

Intros and outros: script it, don't record it

Instead of opening an editor and recording a fresh take, Jordan now writes the intro and outro as a short script — episode title, guest name, one-line hook — and generates the audio using the cloned voice. If a line needs a different pace or emphasis, that's a text edit and a re-generate, not a re-record. A 30-to-45-minute chore became a five-minute step, and every intro sounds consistent regardless of how Jordan's voice actually sounds that particular morning.

Show notes become a bonus audio track

Jordan already writes detailed show notes for search visibility and for listeners who prefer to skim. Now that same text gets pasted into PlayHT's text-to-speech tool and turned into a short narrated audio version in Jordan's own cloned voice — a two-for-one bonus track that goes out to newsletter subscribers and paid members. It used to be extra work nobody had time for; now it's a byproduct of writing Jordan was already doing.

Audiogram voiceovers, in Jordan's own voice, at scale

The biggest shift shows up in social clips. Jordan pulls three or four strong quotes from each interview, writes a one-line setup for each, and generates a short voiceover in the cloned voice to sit under the audiogram waveform. Because generating a voiceover takes minutes instead of a studio session, Jordan went from publishing two clips a week to six or seven, without adding a single hour of recording time.

The results

These numbers are illustrative — every show's numbers will differ based on episode length and format — but they reflect the order of magnitude that Jordan and similar hosts describe after switching:

  • Production time per week: roughly 14–15 hours down to 4–5 hours
  • Social clips published per week: 2 clips up to 6–7 clips
  • Bonus audio content: occasional, unscheduled extras become a narrated show-notes track published with every episode
  • Out-of-pocket cost: no studio time, no freelance voiceover fees, no separate re-recording sessions — just a PlayHT subscription

The reclaimed hours didn't sit idle on a to-do list. Jordan put them toward booking more guest interviews and launching a second, shorter companion show — the kind of expansion that's hard to justify when every new format means more hours in front of a microphone. There's also a quieter benefit that doesn't show up in a hours-saved tally: consistency. Every intro now hits the same energy and pacing, even on weeks when Jordan is jet-lagged, sick, or recording at 6 a.m. before a day job. Listeners never hear the difference, because there isn't one to hear.

A practical tip for podcasters trying this

If you want to try this yourself, don't start with your whole show. Start with the piece you re-record the most, which for most hosts is the intro and outro. Clone your voice once, script those 60 to 90 seconds, and generate them for your next three episodes. If the quality holds up under your own ears, and under your listeners', expand from there into show notes narration and audiogram voiceovers. The workflow compounds: every piece of text you already write — show notes, quote pulls, episode descriptions — becomes a piece of audio you no longer have to record.

Ready to see whether your own voice holds up as a clone? Create a free PlayHT account and test it on your next episode's intro before you decide whether to rebuild the rest of your workflow around it.