Compression & Formats

What Is Transcoding in Audio?

Quick answer

Transcoding means decoding audio from one codec and re-encoding it with a different codec. It's not the same as renaming a file or changing the container. Whether transcoding affects quality depends on the codecs involved — lossless-to-lossless is fine; lossy-to-lossy introduces cumulative quality loss.

Transcoding is not just renaming a file

This is the first thing to understand. Changing a file's extension — renaming audio.mp3 to audio.wav — does not transcode the audio. The file is still an MP3 with the wrong name. Most players will refuse to play it or will play it incorrectly.

Transcoding means actually processing the audio data: decoding it from its current codec representation (turning the compressed bitstream back into raw audio samples) and then encoding it again using a different codec. Two computational steps, not one. The output is genuinely a different format.

Transcoding vs remuxing

Remuxing means moving audio data from one container to another without changing the codec. The audio data itself is untouched — only the wrapper changes. This is fast, lossless, and computationally trivial.

Transcoding changes the codec. It requires decoding (CPU-intensive) and re-encoding (more CPU-intensive). Whether quality is affected depends on what the codecs are.

Remux example:MP4 (AAC audio) → M4A (AAC audio) — container change, codec unchanged. No quality loss.
Transcode example:FLAC → MP3 — different codec. Audio is decoded and re-encoded. Quality changes.

Quality impact by conversion path

Conversion pathQuality impactNotes
Lossless → lossless (FLAC → WAV)NoneBoth sides preserve all audio data
Lossless → lossy (FLAC → MP3)One-time controlled reductionBest starting point for a lossy encode
Lossy → lossless (MP3 → FLAC)None (but no quality gain)Lossless wrapper around lossy audio — bigger file, same quality
Lossy → lossy same codec (MP3 → MP3)Small additional lossDecode and re-encode; artifacts compound slightly
Lossy → lossy diff codec (MP3 → AAC)Noticeable additional lossTwo separate rounds of psychoacoustic removal
Lossy → lossy multiple timesSignificant cumulative lossEach generation degrades further; avoidable

The generation loss problem

Each lossy encode is called a generation. When you re-encode a lossy file to another lossy format, you create a second generation. The degradation compounds: the decoder presents slightly damaged audio to the encoder, which then applies its own round of psychoacoustic removal on top of already-compressed audio.

The artifacts don't double with each generation — but they do accumulate. After three or four lossy re-encodes, the quality has noticeably deteriorated: the high frequencies are smeared, the imaging collapses, the audio sounds muddy or "underwater."

The practical rule: if you ever need to transcode between lossy formats, do it once from the highest-quality lossy source you have. Never use a transcoded file as the source for another transcode if an original lossless version exists.

Best practices to avoid unnecessary transcoding

  • Keep lossless masters:Archive original recordings as WAV or FLAC. Every distribution format can be generated from the lossless master without generation loss.
  • Encode to lossy only once:Generate your MP3 or AAC distribution file directly from the lossless source. If you need a different lossy format, go back to the lossless master, not the MP3.
  • Don't re-encode if remuxing works:If you only need to change the container (MP4 → M4A for AAC audio), check whether the operation is a remux or a transcode. Some tools transcode when they don't need to.

Last updated: March 28, 2026