Podcast Hosting Upload Specs: Buzzsprout, Libsyn, Transistor, Spotify

You spent hours getting the episode to sound exactly right — and then your podcast host quietly re-encoded it. Every major hosting platform applies its own rules to uploads: file-size caps, bitrate transcoding, sometimes a forced conversion to mono. Two podcasters can upload identical files to different hosts and their audiences will hear different audio. This guide consolidates what Buzzsprout, Libsyn, Transistor, and Spotify actually do to your file, and how to prepare an upload that survives each pipeline intact.

Why hosts touch your audio at all

Hosting platforms pay for the bandwidth every download consumes, and they inherited a decade of user uploads at wildly inconsistent settings — 320 kbps music-grade files next to 32 kbps voice memos. Re-encoding everything to a house standard controls costs and protects listeners from broken files. The catch is that re-encoding an MP3 is a lossy-to-lossy generation: if your upload is already close to the house standard the damage is minimal, but if the host transcodes a dense stereo music show down to a speech preset, you will hear it. The defense is simple: know your host's pipeline and deliver a file it will pass through untouched — or one whose transcode is harmless.

Buzzsprout: the re-encoder

Buzzsprout processes every upload to its house format rather than serving your original file — a speech-optimized encode on standard plans, with a higher-quality stereo option available as a paid upgrade, plus optional "Magic Mastering" loudness processing. The practical consequences:

• There is little point uploading 320 kbps — it will be transcoded anyway. A clean 128–192 kbps file, or even a WAV, gives the transcoder the best possible source.
• If your show uses stereo music beds, confirm the stereo option is enabled before your intro collapses to mono.
• Do your own loudness normalization anyway; relying on the platform means your archive copies and cross-posted files stay inconsistent.

Libsyn: storage-quota driven

Libsyn's model meters your monthly upload storage, so file size is money. It serves the MP3 you upload, which cuts both ways: your encoding decisions reach listeners directly, and an oversized file burns quota. A 60-minute episode at 128 kbps runs about 58 MB; at 96 kbps mono it is about 43 MB. On a small plan, that difference decides whether you fit four episodes a month or six. Running finished episodes through an MP3 compressor before upload is the standard Libsyn economy move — the MP3 compression FAQ explains which settings shrink files without audible cost on speech.

Transistor: pass-through

Transistor serves your uploaded file essentially as-is, with generous per-file limits. That makes it the host where your mastering chain matters most: whatever you upload is what every podcast app downloads. The freedom is real but so is the responsibility — an episode accidentally exported at 320 kbps stereo ships as a 144 MB download to every subscriber on cellular. Pass-through hosts are where a pre-upload check of bitrate and loudness earns its keep, and where a bitrate changer fixes an over-encoded episode in one step without reopening the session.

Spotify: the walled garden

Spotify's own hosting (formerly Anchor, now Spotify for Creators) accepts common audio formats and re-encodes for delivery inside Spotify apps, applying its own loudness normalization — around -14 LUFS — at playback. Upload limits are generous enough that any sane podcast MP3 fits comfortably, so the practical guidance is about consistency: Spotify will turn loud episodes down, so mastering hotter than target gains nothing, while quiet episodes get boosted in ways that can raise your noise floor. Deliver the same -16 LUFS file you send everywhere and let the platform adjust playback.

Apple Podcasts: not a host

A common confusion: Apple Podcasts does not store your audio at all. It reads your RSS feed and points listeners at the MP3 sitting on your hosting platform. Whatever your host serves is what Apple listeners hear — so Apple's published guidance on clean encoding and sensible loudness is really a spec for the file you hand your host, not a separate upload.

The universal upload spec

One file satisfies every pipeline above without waste:

• Format: MP3, constant bitrate — the CBR vs VBR FAQ covers why VBR still trips up some players and pipelines.
• Bitrate: 128 kbps stereo for shows with music, 96 kbps mono for pure speech — see the podcast bitrate guide for the reasoning.
• Sample rate: 44.1 kHz.
• Loudness: -16 LUFS stereo or -19 LUFS mono, true peak at or below -1 dBTP.
• Tags: ID3 title, show name, episode number, artwork embedded.

Preparing the file

If your editor exports WAV, encode the delivery file with a WAV to MP3 converter at the settings above rather than accepting whatever default the export dialog last remembered. If you inherited a back catalogue encoded at heavyweight settings, batch it through the bitrate changer before migrating hosts — migration is the one moment you touch every file anyway, and moving 200 episodes at 320 kbps costs real transfer time and storage on the receiving end.

And keep the lossless masters. Hosts change their pipelines every few years — caps move, transcoders improve, loudness targets shift — and the podcasters who re-deliver cleanly are the ones who never made the MP3 their only copy. The spec sheet above fits on an index card, but it is the difference between an episode that sounds like your master everywhere and one that sounds like whatever survived four different transcoders. Upload once, correctly, and every platform's pipeline becomes a pass-through.