Initializing... drag & drop files here
Supports: MP3
AAC (Advanced Audio Coding) is the codec Apple, YouTube, and most streaming platforms standardized on after MP3. It was published as MPEG-2 Part 7 in April 1997 and folded into MPEG-4 Part 3 (ISO/IEC 14496-3) in 1999, using a pure modified discrete cosine transform (MDCT) instead of MP3's hybrid filter bank. The practical result: at the same bitrate AAC encodes more accurately, especially in the high frequencies and at low bitrates where MP3's pre-echo and treble smearing become audible. Common reasons to transcode:
.m4a MP4 container is the recommended audio format for AVFoundation, AVAudioPlayer, and HLS streams. Bundle assets ship smaller and decode with hardware acceleration.| Property | MP3 | AAC |
|---|---|---|
| Standard | MPEG-1/2 Audio Layer III (ISO/IEC 11172-3, 13818-3) | MPEG-4 Part 3 / ISO/IEC 14496-3 |
| Released | 1993 | 1997 (MPEG-2) / 1999 (MPEG-4) |
| Codec algorithm | Hybrid (polyphase + MDCT) | Pure MDCT |
| Max sample rate | 48 kHz (MPEG-1), 16 kHz (MPEG-2) | 96 kHz |
| Max channels | 2 (stereo) | 48 |
| Typical container | .mp3 |
.m4a (MP4) or .aac (ADTS) |
| Patent status | Expired worldwide (2017) | Royalty-bearing for some encoders |
| Default in | Older Windows, generic players | Apple ecosystem, YouTube, HLS |
| Quality at 128 kbps | Good but audibly compressed | Near-transparent for most listeners |
| Use case | Source MP3 | Recommended AAC | How to set it |
|---|---|---|---|
| Spoken word / podcast | 64-96 kbps | 64 kbps mono | Constant Bitrate 64 kbps, Audio Channel Mono |
| Audiobook | 96-128 kbps | 96 kbps mono | Constant Bitrate 96 kbps, Audio Channel Mono |
| General music | 192 kbps | 128 kbps stereo | Constant Bitrate 128 kbps — the preselected value |
| Music library (Apple Music tier) | 256 kbps | 256 kbps stereo | Constant Bitrate 256 kbps |
| Maximum quality from 320 kbps source | 320 kbps | 256 kbps stereo | Constant Bitrate 256 kbps, or the 96k-112k Variable Bitrate band |
| YouTube upload | any | 320-384 kbps stereo | Constant Bitrate 320 or 384 kbps |
No. Both are lossy formats, so transcoding cannot recover frequencies or transients that MP3 already discarded. What AAC does give you is more efficient storage of whatever survived the MP3 encode — a 128 kbps AAC file made from a 192 kbps MP3 will sound very close to the source MP3, whereas a 128 kbps MP3 re-encode would sound noticeably worse. If you still have the original CD rip, WAV, or FLAC, encode AAC directly from that instead.
Constant Bitrate (or Custom Bitrate, if you want a value that isn't in the list). The Quality Preset dropdown displays "Highest" when the page loads, but nothing is pre-selected there — if you never touch it, the server falls back to its own default rather than the value you can see, and the presets are the least predictable control on the page. Name the rate yourself instead, matching it to your source: Constant Bitrate at 256 kbps from a 320 kbps MP3 (perceptually transparent and ~20% smaller), 128-160 kbps from a 128-192 kbps MP3, or 64-96 kbps for voice. Raising the AAC bitrate above the source bitrate wastes space without adding quality. Variable Bitrate is the quality-targeted alternative when consistency matters more than an exact file size.
Variable Bitrate (VBR) is generally better for music — it spends more bits on complex passages and fewer on silence, giving better quality per megabyte. Constant Bitrate (CBR) is preferable when you need predictable file size for streaming chunks (HLS, DASH segments) or when a target hardware decoder is finicky about VBR seeking. For local listening either is fine — the point is to pick one of them explicitly rather than leaving File Compression on Quality Preset, which is the automatic mode and the least predictable control here.
AAC's container (.m4a/MP4) has different overhead than MP3's frame structure, and the encoder may also drop ID3 tags or replace them with iTunes-style metadata atoms. The compressed audio payload at, say, 192 kbps will be very close in size, but the surrounding metadata, padding, and CRC overhead differ.
Yes. Android has decoded AAC natively since Android 3.1 (2011). Windows 10 and 11 play AAC through Groove / Media Player and the Windows Media Foundation decoder. Linux distributions play AAC through GStreamer, FFmpeg, and VLC. Compatibility is effectively universal on devices made in the last decade.
Yes. Expand the Trim section, switch from "Unchanged" to a start time, and set a duration. Times accept seconds or HH:MM:SS.sss with millisecond precision, so you can isolate a 30-second podcast clip or a single chorus without a separate edit step. For longer cutting work see the audio cutter.
.aac and .m4a?Both wrap AAC audio. .aac is the raw ADTS (Audio Data Transport Stream) format — self-synchronizing, used for broadcast and streaming. .m4a wraps the same AAC payload in an MP4 container, giving you metadata, chapter markers, and album art that Apple Music, iTunes, and most music apps expect. This converter outputs the .aac extension; if you specifically need .m4a use MP3 to M4A instead.
If your MP3s already sound fine and you're not on Apple devices, there's no upside to a second-generation lossy transcode. Convert when you need iOS/macOS app compatibility, smaller files at equivalent quality, or HLS-compatible audio. If you ever want to go the other direction (compatibility with older car stereos or generic MP3 players), the reverse path is AAC to MP3.
Three rules: (1) pick the highest-bitrate MP3 you have as the source; (2) set the AAC bitrate at or below the source bitrate — going above wastes space; (3) leave Audio Channel and Sample Rate on Original to avoid an extra resample step. If you have access to the lossless source (CD rip, WAV, FLAC), encode from that with WAV to AAC or FLAC to AAC for genuinely better quality.