How to Transcribe a Podcast or Any Streaming Audio (Spotify, YouTube, Webinars)
Transcription tools want a file. Streaming apps don't give you one — a Spotify episode, a webinar, a video in your browser exists only while it plays. The way around this is older and simpler than it sounds: instead of looking for a file, record the audio your computer is already producing. Here's how to do it properly, when you shouldn't bother, and where the honest limits are.
First: check whether you actually need this
Most podcasts are distributed as plain MP3 files through open RSS feeds — look for a download arrow in your podcast app, or find the episode on the show's own website. If you can get the file, transcribe the file. It is faster than recording in real time, and the audio is pristine rather than re-captured. The technique below is for audio you genuinely cannot download: platform-exclusive shows, webinars, live streams, and browser-only video.
Why there's no file to grab
Streaming platforms deliberately don't hand you a downloadable copy. The audio arrives encrypted or chunked, is decoded on the fly, and is gone as soon as it has played. What they cannot hide, though, is the final step: to reach your ears, the audio has to be decoded and sent to your speakers. At that point it is ordinary sound flowing through your operating system's audio mixer — and your computer can record its own output. That capture is called loopback (or system-audio recording), and it is the same mechanism that lets a transcriber capture both sides of a video call.
Record what you have the right to record
Transcribing an episode for your own notes, research, accessibility, or study is ordinary personal use. Republishing the audio or publishing a full transcript is not — the recording is someone else's copyrighted work, and streaming services' terms of use prohibit copying their content. Keep it for your own understanding, and this stays uncontroversial.
Two ways to do it
Record, then transcribe. Capture the audio to a file while it plays, then run transcription on that file afterwards. This is the accurate route — the model sees the whole recording and you keep the audio for reference.
Transcribe live. Feed the system audio straight into real-time transcription and watch the transcript build as the episode plays. Better when you want to skim or search a talk while it runs; slightly less accurate than a considered pass over a finished file.
In the 360Converter Offline Transcriber these are the Record Audio and Real-Time Transcription features. Both are available on the Premium licence and above, and both run entirely on your machine — the audio is never uploaded.
On Windows
Windows exposes system audio for recording natively (via WASAPI loopback), so nothing extra is needed:
Open the recorder and, in the audio sources, select the remote/system audio (loopback) source. Then — and this is the step people miss — deselect the microphone. You are transcribing the podcast, not your room; leaving the mic on mixes in keyboard noise and background chatter that will end up in your transcript. Start recording, play the episode, and stop when it finishes.
On macOS
macOS does not let applications record system audio on their own, so you need a virtual audio device first. This is a small, standard piece of software that creates a fake audio output your computer can play into and record from — the free, open-source BlackHole is the common choice, and Rogue Amoeba's Loopback is a paid alternative.
After installing it, there's one subtlety worth knowing: if you simply set your system output to the virtual device, the audio goes there instead of your speakers and you'll hear nothing. To both hear and capture, open Audio MIDI Setup and create a Multi-Output Device that includes your speakers and the virtual device, then set that as your system output. Now select the virtual device as the remote/system audio source in the recorder.
Getting a clean result
A few things separate a usable transcript from a frustrating one. Turn the microphone off unless you deliberately want your own commentary in the transcript. Don't mute the playback — loopback records what is actually played, so muting the app or dropping the volume to zero records silence. Set a sensible volume: very low levels record a weak signal, and maxed-out levels can clip. Avoid interruptions — system sounds, notification chimes, and any other tab playing audio all land in the same capture, so silence notifications first.
One expectation to set honestly: a podcast is often two or more people, so you will want speaker diarization to get a readable "who said what" transcript rather than an undifferentiated wall of text. And a recording captured through the speakers is never better than the original file — which is exactly why the first callout on this page says to download the file whenever you can.
What about the video-with-a-file case?
If you already have a video or audio file — a downloaded episode, a recorded lecture, an MP4 from a conference — skip all of the above. Feed the file to the transcriber directly, or read our guide to converting any video or audio into a transcription-ready file if you want to trim, batch, or clean it up first.
Transcribe streaming audio without it leaving your machine
The 360Converter Offline Transcriber records system audio, transcribes it locally, and labels speakers — on Windows and macOS, with nothing uploaded to a server.
Learn more & downloadFrequently asked questions
Can I transcribe a Spotify podcast?
Yes — by recording your computer's system audio while the episode plays, then transcribing that recording. Note that many podcasts on Spotify are also distributed as downloadable MP3s elsewhere; if the episode is available that way, transcribing the file directly gives a better result. Keep the transcript for personal use.
Do I need to install anything?
On Windows, no — system audio capture is built in. On macOS you need a virtual audio device such as BlackHole, because macOS does not expose system audio for recording on its own.
Will it record my microphone too?
Only if you leave the microphone selected. For transcribing a podcast or webinar, deselect it so the capture contains just the streamed audio.
Does the audio get uploaded anywhere?
Not with an offline transcriber. The recording is captured, stored, and transcribed on your own machine — which matters for confidential webinars and internal meetings as much as it does for podcasts.
Which licence do I need?
Audio recording and real-time transcription are included in the Premium, Professional, and Enterprise licences of the 360Converter Offline Transcriber. They are not part of the Standard licence — see License options to compare.
Step details reflect the 360Converter Offline Transcriber as of June 2026; see the Help Center guides to recording mic and remote audio and real-time transcription for the full walkthroughs. Respect the terms of the services you use and the copyright of the material you record.