Six months ago a small podcast producer messaged me in a panic: "Can you turn our latest episode into a transcript, subtitles, and a show-notes page by Friday?" I said yes before I knew how I would do it. I had no paid transcription subscription — no Descript, no Otter, no per-minute billing. What I had was a free Whisper-based transcription tool, a weekend, and the realization that podcasters will happily outsource the busywork they hate. That first gig turned into a repeat client.
Here is the exact workflow I still use to turn podcast audio into a paid freelance line with $0 in software.
Why podcasters pay for transcription
Almost every podcast now needs more than an audio file. Producers want:
- a clean transcript for accessibility and SEO
- burned-in captions for clips posted to social video
- a show-notes page that turns a casual listener into a subscriber
Do that by hand and one hour of audio costs four to six hours of typing. Most podcast hosts are creators, not typists — so they outsource. The work is repetitive, deadline-driven, and exactly the kind of thing a busy producer will pay someone to remove from their plate.
The money adds up: small interview-style shows pay per episode, while corporate podcasts and YouTube channels with several uploads a week become monthly retainers. Because every new episode produces the same deliverables, a happy client reorders constantly.
The $0 stack I run the whole service on
You do not need paid software to start this. My entire stack is free or runs locally.
1. Whisper for the raw transcript
Whisper is still the gold standard for accuracy, handling timestamps, punctuation, and speaker turns well in 90+ languages. I run it through a free open-source desktop interface so client audio never leaves my machine — no uploads, no privacy leak, no account wall. Ten hours of interviews transcribe while I sleep.
2. A text pass for structure
Raw Whisper output is a wall of text. Before I deliver, I split it into paragraphs, label speakers, and flag timestamps for every topic. Free apps handle the cleanup; my time goes into the judgment calls software cannot make — who said what, where a section should break, which quotes deserve highlighting.
3. Subtitle export to SRT/VTT
Clients increasingly want captions they can drop straight into YouTube, TikTok, or a video editor. A good transcription tool exports subtitle files directly, so I deliver a ready-to-upload SRT alongside the transcript instead of making the client rebuild captions by hand. That one extra file is why clients stay: they are paying for finished deliverables, not homework.
4. Show notes built from the transcript
Because I already have clean text with speaker labels and timestamps, writing show notes is a ten-minute job. Timestamped topic links, a two-line summary, and a few highlight quotes turn a raw episode into a page that ranks and converts. Freelancers who skip this leave easy money on the table.
What I charge and where I find clients
Pricing depends on how finished the client wants the work:
- $30–$60 per episode hour for simple transcripts
- $80–$200 for a polished one-hour episode with transcript, captions, and show notes
- monthly retainer for ongoing multi-episode producers
The range is wide precisely because the deliverables are — scope clearly in the proposal and you never argue about price later.
I find clients where podcasters already hang out: podcast subreddits and communities full of first-time hosts, agencies producing branded shows, YouTube channels that post hour-long interviews with no captions, and local business podcasts that need transcripts for SEO. A short portfolio of three sample episodes — "I turned this 40-minute interview into a transcript, captions, and show notes in an afternoon" — closes most conversations.
Honest ceiling
Be straight about the limits before you commit. Raw transcription is a commodity, so the margin lives in the finished bundle — structure, captions, and show notes — not in the raw text. Accuracy tools still miss accents, music beds, and heavy cross-talk, so you will edit. And client volume is real but rarely explosive; this is a dependable side income and a solid first freelance offering.
FAQ
Do I need any coding skills to run this?
No. Free Whisper desktop apps have a graphical interface — you drop in an audio file and get a transcript back. The skills that matter are cleanup accuracy and reliable delivery, not programming.
Is this legal and above board?
Yes, as long as you transcribe audio the client owns or has rights to, and you do not resell the tool itself. You are selling your labor and formatting, which is straightforward freelance work.
How do I start with zero clients?
Offer two or three small local or community podcasters a discounted first episode in exchange for a testimonial and a sample you can show. A tiny proof-of-work portfolio outsells any pitch.
This article first appeared on my blog. You can find the original and a growing collection of $0 creator workflows there.
Top comments (0)