Is transcoding loss actually audible?

It depends on the quality of your source. Converting a 256 or 320 kbps MP3 to AAC at the same level produces a difference too small to distinguish in most listening environments. Dropping a 128 kbps MP3 to 96 kbps AAC, on the other hand, makes the degradation clearly audible. The better the source, the less the cost of the second encode is felt.

Why don't the two encoders make the same decisions?

MP3 and AAC use different psychoacoustic models, different block sizes, and different approaches to frequency partitioning. The AAC encoder discards what its own model considers unnecessary, not what MP3 already discarded. Because those two sets of decisions do not overlap, the second pass also removes some of the information that survived the first. The losses add up rather than cancelling out.

Is there a way to eliminate the loss entirely?

If you start from a lossy source, no, it cannot be reduced to zero. The best you can do is keep the output bitrate equal to or one step above the source and keep the chain short. The only way to genuinely avoid the loss is to convert from a lossless source such as WAV or FLAC, in which case only a single encoding pass occurs.

Does raising the bitrate compensate for the loss?

It does not compensate, but it does reduce it. A high bitrate gives the second encoder a budget generous enough to fit the audio comfortably, lessening its need to discard new information. However, this does not recover what was lost in the first pass — it merely limits the extra damage from the second. The file grows, but the audio never becomes better than the source.

The Double Compression Problem in MP3 to M4A Conversion

A short definition of the problem

You have an MP3 and you want to convert it to M4A. The operation takes a few seconds and the file opens without complaint. But something irreversible has happened in your audio chain behind the scenes: the loss occurred twice.

This is called transcoding loss, or generation loss. It appears in every conversion that goes from one lossy format to another, and its effect is cumulative. This article explains why it is unavoidable, how much it actually matters, and how to suppress it in practice.

How lossy compression works

Lossy encoders such as MP3 and AAC do not store audio as it is. Instead they use a psychoacoustic model: it predicts which components the human ear will fail to notice and removes those components from the file.

For instance, a very quiet detail arriving immediately after a loud drum hit is inaudible anyway — the encoder discards it. High frequencies above the hearing threshold are cut too. The remaining data is then quantised in coarser steps, meaning each value is stored only approximately.

All three operations share one property: they destroy information. There is no backup inside the compressed file that could restore the deleted portion.

Why does the second pass cause extra damage?

This is the crucial part. Converting MP3 to M4A happens in two stages:

Decoding. The MP3 file is expanded into raw PCM samples. At this point you hold a waveform that already carries the losses of the first encode.
Re-encoding. That waveform is handed to the AAC encoder, which compresses it again according to its own model.

Here is the problem: the AAC encoder has no idea the audio it received was compressed before. As far as it is concerned, this is an ordinary recording. It applies its own psychoacoustic model and discards whatever its criteria deem unnecessary.

But the two encoders' decisions do not overlap. MP3 and AAC use different block sizes, different frequency partitioning approaches, and different masking thresholds. AAC discards not the places MP3 already discarded, but the places its own model considers removable — and some of those are things MP3 had preserved.

The result: losses accumulate rather than cancelling out.

An additional problem: encoding the artefacts

There is a subtler issue too. MP3 encoding, particularly at low bitrates, leaves small artefacts behind: slight pre-echo, a metallic texture in the high frequencies, softening on sharp transients.

The second encoder sees those artefacts as "audio" and tries to preserve them as well. In other words, it spends part of its budget storing distortions that are not actually music. That reduces the data available for genuine sound and lowers efficiency.

This is why converting an already low-bitrate MP3 can degrade disproportionately.

How much does it matter? Let us be realistic

Read this far and the picture may look bleak. In practice, though, the situation is usually tolerable — and that deserves saying honestly too.

If the source is good, the difference is very small. Converting a 256 or 320 kbps MP3 to AAC at the same level produces a difference most people cannot distinguish under normal listening conditions. On a phone speaker, in a car, or in a noisy environment, it disappears entirely.

If the source is weak, the difference grows. A 128 kbps MP3 is already at the edge; dropping it further to a lower AAC bitrate makes the degradation plainly audible.

It worsens as the chain lengthens. A single conversion is generally not a problem. But repeated conversions such as MP3 → M4A → MP3 → OGG take a little more each pass, and the effect accumulates.

Practical rules for minimising the loss

1. Start from a lossless source if you can

This is more effective than every other recommendation combined. If a WAV or FLAC version of the same recording exists, convert directly from it. Only a single encoding pass occurs, and you genuinely benefit from AAC's efficiency advantage.

2. Keep the bitrate equal to or one step above the source

Encode a 192 kbps MP3 as 192 or 256 kbps AAC. A higher bitrate does not improve quality — that needs stating plainly. What it does is give the second encoder a comfortable budget, reducing its need to discard fresh information.

3. Keep the chain short

Every conversion pass has a cost. Do not bounce between formats unnecessarily; reach your target format in as few steps as possible.

4. Ask whether it is genuinely necessary

This is the most commonly skipped step. You do not need to convert MP3 to M4A in order to listen on an iPhone; the iPhone plays MP3 without issue. Conversion only makes sense with a concrete need such as a specific app, ringtone creation, or an audiobook workflow.

5. Keep the original

If you chose the wrong setting or dislike the result, you can try again as long as the source remains. Delete it and all you have left is a twice-compressed copy.

Applying these rules, you can make informed choices in our MP3 to M4A converter.

In summary

Converting a lossy file into another lossy format always carries an extra cost, because two encoders discard different information and those losses accumulate. This is unavoidable, but it is manageable: choose a bitrate equal to or slightly above the source, keep the chain short, and if you have a lossless version, start from it. Most importantly, convert for the right reason — compatibility and workflow are valid ones, while gaining quality is an expectation that will never be met.