Convert MPG to AIFF Online for Free

MPG to AIFF Extracts Audio From a Timed Programme

An MPG file is commonly an MPEG Program Stream. It may hold video plus one or more audio elementary streams, not one ready-made audio recording. Program Streams assemble variable-length packetized elementary stream (PES) packets into packs, and PES packets can carry presentation timestamps that synchronize sound with picture. AIFF is audio only. The conversion selects a soundtrack, decodes it into PCM samples, and writes those samples into an AIFF file; it does not retain video, captions, navigation, or unselected languages.

Inspect tracks before extraction. Record codec, language/role, channels, sample rate, duration, and whether the track is dialogue, commentary, description, music, or an alternate dub. Listen at an opening cue, a middle landmark, and the final tail. A first listed stream is not necessarily the desired one. Keep the MPG source because it remains the only record of picture context, other tracks, source packet timing, and later delivery choices.

AIFF is appropriate when a receiving editor, Apple-oriented audio workflow, or uncompressed interchange requirement explicitly asks for it. It is not a way to retrieve detail lost by earlier MPEG Layer audio, AAC, or another lossy source codec. The decoded PCM represents the best accessible rendering of that selected source stream, not a return to any earlier studio master.

AIFF Stores PCM in Named FORM, COMM, and SSND Chunks

AIFF is a chunked audio format. Its outer form begins with the identifiers FORM and AIFF. The Common Chunk, COMM, describes the number of channels, sample frames per channel, bits per sample, and sample rate. The Sound Data Chunk, SSND, holds the sample data and its offset/block information. Apple describes AIFF as a cross-platform format whose Common Chunk carries sample-rate and bit-depth information and whose Sound Data Chunk contains the actual samples.

That structure is much simpler than a video programme, but it does not carry the whole MPG context. AIFF can describe the extracted audio sample stream; it does not preserve MPEG video frames, PES identifiers, captions, a movie-style selection menu, or the reason a particular track was chosen. Copy title, creator, language, date, rights, and artwork only when verified. A filename or an on-screen credit is not reliable proof of authorship or permission.

Use standard uncompressed AIFF when that is what the receiver requests, and distinguish it from AIFF-C where a compressed form is expected. A renamed MPG file contains neither AIFF chunks nor decoded PCM. A valid result should expose the expected sample rate, channels, bit depth, and duration to the receiving application.

Lossless AIFF Begins Only After the MPG Audio Decode

Uncompressed PCM AIFF does not apply a new perceptual audio codec after decoding. It can therefore avoid adding another lossy audio stage. That is useful for editing or a receiver that specifically needs PCM, but it must not be described as restoration. If the MPG soundtrack was already compressed, high-frequency detail, stereo cues, transients, or noise characteristics altered by that source codec are already part of the decoded result. AIFF preserves the resulting samples; it does not recreate missing original data.

• Gain: decoded samples can be delivered without a further lossy audio encode.
• Loss: video, captions, track menus, and omitted languages stay with the MPG.
• Size result: uncompressed PCM size rises directly with duration, sample rate, channels, and bit depth.
• Rate rule: resampling may satisfy a receiver but cannot create missing source bandwidth.
• Channel rule: downmixing surround changes balance and phase; duplicating mono does not recover stereo.

For uncompressed PCM, estimated payload size is duration multiplied by sample rate, channels, and bytes per sample. A one-hour stereo 48 kHz, 16-bit stream has roughly 691,200,000 audio bytes before chunks and metadata: 3,600 × 48,000 × 2 × 2. This explains why AIFF is much larger than a lossy MPG soundtrack even when it does not improve the historical audio quality.

Programme Timing Determines the Audio Boundary

MPG presentation timing belongs to the programme. PES streams in a Program Stream share a clock, and a PES header can carry a presentation timestamp. Decide whether AIFF should contain the sound as presented with picture, the full decoded audio stream, or a precisely named excerpt. Do not decide from a byte count. Leading silence may be intentional alignment to picture; an arbitrary trim can remove an opening consonant, drum hit, or reverberation tail.

Compare first audible sound, a middle landmark, and final tail. A fixed shift points to an initial offset or trim policy. A difference that grows through the programme points more often to source timestamps, sample-rate mapping, or damaged input. AIFF does not automatically repair either fault. If a source reports a corrupt or missing PES packet, retain that warning: a smooth-sounding PCM output cannot prove the original packet was present.

Record the selected source track, source codec/layout, start/end policy, output sample rate/channels/bit depth, output duration, metadata copied, and target tested. That record is essential because an audio-only derivative cannot show which of several MPG tracks it came from.

AIFF Compatibility Is Not the Same as General Audio Support

Test AIFF in the named application or device, not merely a convenient player. Apple documents AIFF as an audio format, while Microsoft now says support for AIFF/AIF/AIFC and MPEG-2 media is limited and deprecated in PowerPoint version 2505 and later. That specific example shows why a file that works in an audio editor may still be unsuitable as a presentation-media handoff. The receiver's current version and accepted PCM properties decide the delivery choice.

Check opening, displayed duration, channel interpretation, sample rate, audio at three landmarks, and final playback. A narrower receiver may expect stereo, a particular sample rate, or ordinary uncompressed AIFF instead of a variant. If AIFF is going into an editing system, also test import and export; playback alone does not prove the application will preserve its properties. Keep the original MPG for any receiver that later needs video or a different soundtrack.

AIFF metadata can be helpful but should be treated as descriptive, not as a substitute for a manifest. Preserve a separate note identifying source file, selected track and language, timing choices, source warnings, output properties, and verification result.

Common MPG Audio Extraction Errors Have Specific Causes

Wrong language or commentary in the AIFF means the wrong elementary audio stream was chosen. Inventory and audition all tracks before the write. A clipped start or end means the timeline policy, PES timing, or encoder/decoder handling needs inspection; do not solve it by blindly adding silence. A stretched duration or growing sync mismatch needs a source time/rate investigation rather than a different extension.

An AIFF that is unexpectedly huge is often correct: raw PCM does not reduce silent or simple passages in the way lossless compression can, and size is determined largely by the four values in the size calculation. An AIFF with unexpected quality loss was likely made through an unnecessary lossy intermediate or a downmix/resample choice. Return to the selected MPG stream and make the intentional conversion once.

If a receiver rejects the file, inspect its requested container, maximum channels, sample rate, bit depth, and whether it accepts uncompressed AIFF. Do not relabel a different format as AIFF. A real AIFF must have coherent form/chunk structure and sample data consistent with its Common Chunk values.

MPG soundtrack and AIFF PCM comparison:

Questions Before Exporting AIFF From MPG

Does AIFF keep the video?
No. AIFF is audio-only; retain the MPG for picture, captions, and programme context.

Does AIFF restore lossy MPG audio?
No. It preserves decoded PCM without another lossy audio encode but cannot recover earlier loss.

Why is the AIFF much larger?
Uncompressed PCM stores samples directly, so size follows duration, sample rate, channels, and bit depth.

Why is the wrong language audible?
An unintended MPG audio stream was selected. Inspect and audition tracks first.

Should MPG be deleted after extraction?
No. Keep it for alternate tracks, source timing, video, captions, and later conversions.