Think AAC instead of MP3

2

In blind tests, AAC demonstrates better sound quality than MP3, and has the advantage of being an international standard with playback on virtually every modern device.

The first MPEG layer-III mp3 encoder, l3enc, was released in July, 1994 – over a quarter century ago. While it seemed advanced by early 90’s standards, better encoding algorithms exist today. More recently engineered codecs account for the psychological limitations of human hearing, and bias data compression to preserve more audible information in the same space or less than an MP3. 

The MPEG layer-III codec initially became popular after its implementation as an international standard in the late-90’s by the Motion Picture Experts Group (MPEG). By 1998, MPEG introduced Advanced Audio Coding (AAC) that allowed for more sample rates, more channels, and higher coding accuracy compared to MP3. In blind tests, AAC demonstrates better sound quality than MP3, and has the advantage of being an international standard with playback on virtually every modern device. 

Content creators and distributors still tend to use MP3 due to its familiarity, even at the cost of additional data and poorer performance. Other efficient audio codecs exist, such as the open-source formats, Opus and Speex, but these lack the widespread compatibility of AAC.

The good news is that encoding to AAC is incredibly simple on almost every audio and video editing software. In this post, we’ll discuss AAC encoding settings for video streaming, music and speech. 

Note that not all audio needs to be encoded into a lossy format. For high quality or archival-grade audio, we recommend using PCM, FLAC, or ALAC, which preserve all of the original audio data without loss. For non-archival audio data intended for streaming or for reducing disk storage, AAC is one of the best sounding and most efficient codecs available.

AAC Settings For H.264 Video

When exporting H.264 video for distribution, you’ll have configure the audio bitrate and coding settings upon export. Generally speaking, stereo AAC audio at 192kbps is virtually indistinguishable from lossless PCM audio and suitable for any kind of stereo audio playback. For video that contains only dialogue and speech without music, export as mono and compress the audio to 96kbps for very high-quality, or even further to 64kbps where file size is a concern. 

However, for uploading videos to streaming services that compress the audio data even further, it’s best to use a data rate of 320kbps. As of January 2020, streaming services like YouTube and Vimeo list 320kpbs as their upper audio data limit for H.264 video uploads. Providing the largest amount of data allows for better quality to the severs before applying an additional pass of data compression when the streaming service sends the data to its users. For 5.1 surround sound, default to 384kbps or greater. 

AAC Settings For Music

Sending smaller audio files is becoming increasingly common for artists, producers, and mix engineers. We’ll often record live music sets or email them as attachments to the performers using email services like Gmail,  with attachment limits of 25MB. If necessary, save a lossless version as an archive, but use AAC copies to send to the artists or to keep on local storage for playback. (Of course, we recommend sending the artist a lossless file if requested!)

Whether sending music over the web or storing copies of audio files, 128kpbs stereo AAC will more than suffice for any genre of music. For the discerning listener, use an even higher data rate of 256kpbs for virtually flawless recordings. Feel free to go lower to roughly 100 kpbs when the additional space is needed, as the compression artifacts will still be barely noticeable. 

Compressing a high-quality file for music can be accomplished easily in MacOS. Start with your a high-quality audio file(s) in Finder, then right-click the file and select “Encode Selected Audio Files”. 

In the “Encoder” dropdown box, select “High-quality” for 128kpbs AAC, or iTunes Plus for an even higher quality 256kpbs AAC file. 

Next, click “Continue” and finder will compress the audio file to the more web-friendly AAC format. 

AAC Settings For Speech and Dialogue

Whether it’s narration, dialogue, a podcast with little to no music, or even a voice memo, audio containing only human voices should be exported in mono at 64kpbs. AAC is remarkably efficient at encoding high quality audio to containing only the spoken vocals, since human voices mostly contain midrange frequencies that require less data to encode. For audio files where sound quality isn’t a concern, feel free to drop the bitrate even further to 28kbps for perfectly clear and intelligible recordings, albeit with some minor data compression artifacts. 

If uploading to a streaming service, default to 192kbps mono or consult the service’s submission guidelines for the maximum upload data rate. 

Compressing a high-quality file for spoken vocals can be accomplished easily in MacOS. Start with your high-quality audio file(s) in Finder, then right-click the file and select “Encode Selected Audio Files”. 

In the “Encoder” dropdown box, select “Spoken Podcast” for a small, variable bitrate AAC file. 

Next, click “Continue” and finder will compress the audio file to the more web-friendly AAC format. 

Summary

Consider using AAC rather than MP3 for lossy encoding, as it provides better audio quality at the same data rates or lower than MP3. You’ll be able to save storage space and send data more efficiently, at the same audio quality or better. If you have questions or found this helpful, let us know in the comments below!

Photo Credit: Eryk Thompson

CEO and Founder of Finewav. Instagram: @ErykThompson Vero: @ErykThompson

Related Posts

Leave a Reply

My New Stories

Save this space with lossless compression by using audio codecs like FLAC and ALAC