Initializing... drag & drop files here
Supports: MPEG2
An .mpeg2 file is an MPEG program stream, so the audio you want is already in there beside the video — usually MPEG-1 Layer II or Dolby AC-3 from a DVD rip or a broadcast capture. This tool strips the picture away and re-encodes just the soundtrack as AAC inside an MPEG-4 audio container, which is the format iTunes, Apple Music, Books and every modern phone treat as native. AAC is the only output codec on this page, so the decisions worth making are about bitrate, channels and how much of the timeline you keep.
.mpeg2 file onto the page or click "+ Add Files". Several files can be queued and processed with one set of settings.HH:MM:SS.sss if you only want one segment..m4a. Files are uploaded over an encrypted connection, processed on our servers, and deleted automatically after a few hours — no sign-up, no watermark.The Variable Bitrate entries are AAC's own quality modes. Read them as an ordered quality scale rather than as arithmetic: the encoder's documentation describes the numbers per channel, but measured against real audio the total rate barely moves with channel count and the top rung lands well above its printed label.
| File Compression mode | What you set | Stereo result, roughly | Use it when |
|---|---|---|---|
| Variable Bitrate | One of five quality steps, labelled 20k-32k up to 96k-112k | Rises with each step; the top step measured around 150 kbps | You care about consistent quality and don't need an exact size |
| Constant Bitrate | A fixed rate from 8 to 384 kbps (128 kbps preselected) | Exactly what you picked | A player or device expects one predictable rate |
| Custom Bitrate | Any value you type, in bps / Kbps / Mbps | Exactly what you typed | You need a rate the dropdown doesn't list |
| Specific file size | A target in bytes, KB or MB | Whatever fits | You have a hard upload or storage ceiling |
Naming one of these explicitly is what gives you a result you can plan around. File Compression opens on the Quality Preset row, which is the automatic, hands-off mode: nothing is pre-selected in that dropdown, so leaving it untouched hands the decision to the server's own default rather than the "Highest" on screen. It is also the least predictable control on the page, so when the output rate matters, use one of the four bitrate modes above instead.
| Target | Codec | Lossy? | Best reason to pick it |
|---|---|---|---|
| M4A (this page) | AAC | Yes | Apple ecosystem, audiobooks, best quality per kilobyte at mid bitrates |
| MP3 | MP3 | Yes | Maximum compatibility — car stereos, old players, anything with a USB port |
| WAV | PCM | No | Editing in a DAW; very large files |
| FLAC | FLAC | No | Archiving the extracted track without further loss |
| OGA | Vorbis / Opus | Yes | Open, royalty-free formats for desktop and web playback |
The one with the most channels, so on a typical DVD program stream the 5.1 mix wins over the stereo mix. There is no track picker on this page: when several audio streams are present the converter falls back to the default rule, which selects the stream with the greatest channel count and breaks ties by choosing the lowest stream index. If that gives you a downmixed-from-5.1 result you didn't want, setting Audio Channel to Stereo at least gives you a predictable two-channel output rather than an untouched surround layout.
Because the labels are approximate rather than exact. The encoder's documentation describes them per channel, which would put the top entry near 200 kbps for stereo — but encoding the same source as mono and as stereo produced almost the same overall rate, and the top entry measured around 150 kbps either way. Treat the five steps as an ordered quality scale where higher is better, and use Constant Bitrate or Custom Bitrate when you need the number you enter to be the number the file gets.
No. The audio in an MPEG-2 program stream is already lossy — MPEG-1 Layer II and Dolby AC-3 both discard information at record time — and re-encoding it as AAC is a second lossy pass. A generous bitrate keeps the result close to the source, but nothing recovers what the first encode threw away. The gain here is convenience and file size, not fidelity. If you want to avoid a second lossy generation entirely, take the track to FLAC or WAV instead.
Leave Audio Sample Rate on "Original" unless you have a reason not to. DVD and broadcast audio is typically 48000 Hz, and resampling it to 44100 Hz gains nothing while adding a conversion step. The lower entries (8000 to 24000 Hz) exist for speech-only material where a small file matters more than bandwidth — a lecture or interview capture, for instance. Never resample upward: it enlarges the file without adding any high-frequency content.
.m4a different from .mp4 or .aac?.m4a and .mp4 are the same MPEG-4 container; the .m4a spelling is the convention for a file that holds only audio, and it is what makes Apple software treat it as music rather than video. A raw .aac file is different — that is a bare AAC stream with no MPEG-4 container around it, so it cannot carry cover art, chapters or reliable duration metadata. Use .m4a unless something specifically demands the bare stream, in which case MPEG-2 to AAC produces it.
Yes, with the Trim control. Set a Start time and a Duration, in plain seconds or in HH:MM:SS.sss form, and only that span is encoded. That is the efficient route for pulling one song or one scene out of a long capture. Note that Audio Cutter does not accept video files, so if you want to cut before converting, use Video Cutter on the MPEG-2 source instead.
No. Those live in the DVD or program-stream structure, not in the audio elementary stream, and none of them survive an audio-only extraction. What you get is the sound. Any chapters or artwork you want in the finished .m4a have to be added afterwards in a tagger such as MusicBrainz Picard or the metadata editor in your player of choice.
No — AAC is a lossy codec at every bitrate, including the highest one the Constant Bitrate dropdown offers. What a very high rate buys you is that the second generation of loss becomes hard to hear, not that it stops happening. Above roughly 256 kbps for stereo you are mostly spending disk space. If the goal is genuinely no further loss, the lossless targets are the right answer instead.
It is uploaded over an encrypted connection, processed on our servers, and both the uploaded video and the generated .m4a are deleted automatically after a few hours. No account is needed, nothing is watermarked, and files are never shared or made public. On a multi-gigabyte DVD rip the slow part is the upload rather than the encode, so trimming the source first is the fastest route when you only need a few minutes of audio.