Initializing... drag & drop files here
Supports: MP4, M4V
MP4 is a container that almost always carries AAC audio alongside H.264 or H.265 video — great for playback on phones and TVs, but oversized when you only need the soundtrack. MP3 is a lossy audio-only format that has remained universally supported by car stereos, MP3 players, smart speakers, DAWs, podcast hosts, and every operating system since the late 1990s. Extracting the audio drops file size by 90-95% and lets the audio travel anywhere video can't.
| Property | MP4 (audio track) | MP3 |
|---|---|---|
| Typical codec | AAC-LC (sometimes AAC-HE, AC-3, or Opus) | MPEG-1 / MPEG-2 Layer III |
| Audio + video together | Yes | No, audio only |
| Standard bitrate range | 96-256 kbps stereo (AAC) | 8-320 kbps (CBR) or VBR V0-V9 |
| Quality at 128 kbps | Roughly equal to MP3 at ~192 kbps | Baseline "good enough" for speech |
| Universal device support | Modern devices only | Every audio player since 1998 |
| Tag support | iTunes-style metadata in moov atom |
ID3v1 / ID3v2 (cover art, lyrics) |
| Royalty status | Patent-free since 2017 (MPEG-4 base) | Patent-free since 2017 |
| Typical 1-hour size | 800 MB - 2 GB (with video) | 55-150 MB (audio only) |
| Bitrate | File Size | Best For | Notes |
|---|---|---|---|
| 320 kbps CBR | ~140 MB | Music videos, concerts, mastering | MP3's maximum; transparent for most listeners |
| 256 kbps CBR | ~115 MB | Hi-fi sharing, archival | Diminishing returns vs 320 kbps |
| 192 kbps CBR | ~85 MB | General listening, mixed content | Most common "good quality" setting |
| 128 kbps CBR | ~55 MB | Podcasts, interviews, lectures | Reference quality for speech; standard for older streaming |
| 96 kbps CBR | ~42 MB | Voice memos, audiobooks | Noticeable artifacts on music |
| 64 kbps CBR | ~28 MB | Mono speech, low-bandwidth | Acceptable for voice only |
| VBR V0 (~245 kbps avg) | ~110 MB | Music when size matters | LAME quality target; near-CBR-320 quality |
| VBR V2 (~190 kbps avg) | ~85 MB | Best size-to-quality tradeoff | Recommended default for mixed content |
| Source type | Recommended bitrate | Channels | Sample rate |
|---|---|---|---|
| Music video, concert, DJ set | 256-320 kbps CBR or VBR V0 | Stereo | 44100 Hz |
| Podcast interview (single mic) | 96-128 kbps CBR | Mono | 44100 Hz |
| Podcast (multi-host, music intros) | 128-192 kbps CBR | Stereo | 44100 Hz |
| Lecture, webinar, Zoom recording | 96-128 kbps CBR | Mono | 44100 Hz |
| Audiobook narration | 64-96 kbps CBR | Mono | 44100 Hz |
| Mixed YouTube content | 192 kbps CBR or VBR V2 | Stereo | 44100 Hz |
For the reverse direction or other targets, see MP3 to WAV, MP4 to WAV (lossless), MP4 to FLAC, MP4 to AAC, or MP4 to M4A. Already have an MP3 that's too big? Try Compress MP3.
The MP4 almost certainly contains AAC, which is already lossy. Going AAC to MP3 is a transcoding step — there is some quality loss, but at 192 kbps MP3 or higher most listeners cannot tell the difference from the source on consumer speakers or earbuds. The source bitrate is the ceiling: re-encoding a 128 kbps AAC track at 320 kbps MP3 does not add detail back; it only inflates file size.
YouTube's stereo AAC audio for 1080p uploads is typically 128-192 kbps. Encoding to MP3 at 192-256 kbps captures everything the source has to offer; 320 kbps wastes space without adding fidelity. For a 1080p60 or 4K source with stereo Opus/AAC, 256-320 kbps MP3 is a safe choice if you want maximum headroom.
File size depends on bitrate and duration, not on the original video size. A 1-hour MP4 at 1080p might be 2 GB, but a 1-hour 192 kbps MP3 is always close to 85 MB regardless of whether the source was 720p or 4K. If your output looks tiny, the source audio may have been very short or mono — check the trim settings and channel dropdown.
Yes — use the Trim controls under Advanced Options. Set the Start point (HH:MM:SS.MS format) and Duration to pull a single song from a long concert MP4, a 30-second hook for a ringtone, or just the Q&A section of a webinar. For more advanced cutting after conversion, use the MP3 trimmer; to trim the video first, see the video trimmer.
Constant Bitrate gives predictable file sizes and broadest legacy hardware compatibility (older car stereos, hardware MP3 players), which is why streaming and podcast platforms still default to it. Variable Bitrate produces smaller files at the same perceived quality by allocating more bits to complex passages — best when you want maximum quality per megabyte and your playback device is from the last decade.
The most common cause is the audio stream uses an unusual codec or the file uses a fragmented MP4 (fMP4) variant from a DRM-protected stream. Files downloaded from streaming services may carry encrypted audio tracks that no converter can extract. Try re-saving the file with a desktop tool first, or upload an MP4 that plays normally in VLC or QuickTime.
Basic metadata (title, artist, album if present in the MP4) is mapped to ID3v2 tags on the MP3. Cover art embedded in the MP4 moov atom is preserved when present. MP4 chapter markers do not have a clean MP3 equivalent — MP3 supports chapter frames via ID3v2 but most players ignore them; if chapters matter, M4A (AAC) is a better target.
Extracting audio from a video you own or have rights to use is fine. Pulling audio from copyrighted music videos, paid streaming services, or DRM-protected content for redistribution is not — most jurisdictions treat that as a copyright violation. Personal-use ripping rules vary by country (e.g., permitted in the US under fair use in limited cases, restricted in Germany, allowed for private copies in France with levies). Check the rules where you live.
Audio extraction reads the audio track at its native timestamps, so any drift baked into the original MP4 will carry over. If the source video had visible lip-sync issues, the MP3 will start at the same absolute time but won't "fix" the drift. For files with bad timestamps, re-mux the MP4 with ffmpeg or a desktop tool first, then extract.