How to transcribe a podcast episode for free
A practical podcast transcription workflow: get the right audio file, generate a timestamped first pass, verify names and quotes, then prepare show notes and subtitles.
Start with an audio file you have the right to use
The best source is the original recording or the file on your own podcast hosting dashboard. A Spotify or Apple Podcasts page is not an audio file, so a browser tool cannot turn the page URL directly into text. Do not rip or download someone else's episode unless you have permission. For a show you own, export an MP3, M4A, WAV or compatible video file.
Decide what the transcript must produce
A full episode transcript is useful for search, accessibility and future repurposing. Show notes answer a different question: why should someone listen? Subtitles need short, timed captions. Decide whether you need one output or all three before you start, because each one needs a different amount of cleanup.
Generate a timestamped first pass
Upload the episode and let the speech model produce a timestamped transcript. Choose the spoken language when auto-detection is uncertain. For long files, confirm that the tool processes segments in order and exposes recoverable progress instead of making you restart the whole episode after one failed request.
Verify the details an automatic transcript can miss
Check names, organizations, product names, numbers, dates, links and every quote you plan to publish against the audio. Pay particular attention to accents, crosstalk and technical vocabulary. A fluent-looking transcript can still be wrong. The transcript is a first draft and a searchable map into the episode, not a human-certified record.
Turn the first pass into useful show notes
Read the transcript once for structure, then write a one-sentence episode summary, three to five specific chapters, two verified quotes and the single next step you want a listener to take. Use the timestamp to confirm each quote. Do not paste the entire transcript into the show-notes field and call it finished.
Export subtitles in the format your editor accepts
SRT is widely supported by video editors, YouTube and social tools. WebVTT is common on the web. Check line length and reading speed after export, and correct speaker changes that the model did not identify. Automatic subtitles still need a visual review before publication.
Understand the practical cost of a free workflow
Loqena currently provides 30 free transcription minutes per day, capped at 180 minutes per month. A one-hour episode may therefore take more than one free session. If you need the whole episode immediately, the one-time 60 Minute Pack is $5 and does not renew. A recurring Pro plan is only worth choosing if you publish enough episodes to use the included minutes each month.
Keep private recordings under control
Only upload audio you are allowed to process. Use public clips for product tests and review your organization's requirements before sending confidential interviews to any cloud transcription service. Keep the original recording in your own storage and export only the transcript files you actually need.