Audio formats and bitrates explained

Step by step

1. Work out what you have. Check the file extension and, on a Mac, press Cmd+I; on Windows, right-click, Properties, Details. You will see the format, bitrate and sample rate.
2. Decide what the file is for. Listening on a phone, uploading to a podcast host, sending a voice note, editing in a DAW, or archiving. Each points to a different format and bitrate.
3. Choose the format from the table. MP3 at 192 kbps for general listening and sharing, M4A for the best size at a given quality, WAV for editing and mastering, FLAC for archives.
4. Convert in the browser. Drop the file on WAV to MP3, M4A to MP3, MP3 to WAV or MP4 to MP3, pick the MP3 bitrate or sample rate, and click Convert. Nothing is uploaded.
5. Pull audio from video without loss. Open Extract Audio, choose M4A (copy original AAC, instant) and click Extract audio. The AAC stream is copied straight out of the video, so there is no re-encoding and no quality loss.

What does bitrate actually mean?

Bitrate is the number of bits used to describe each second of sound, written in kilobits per second. Multiply by the length and you have the file size: 192 kbps for a 4 minute song is 192,000 times 240 seconds, about 46 million bits, or 5.8 MB. A higher bitrate gives the encoder more room to keep detail, so quality rises with size, but with diminishing returns. Lossy codecs such as MP3 and AAC spend their bits on the parts of the sound the ear notices and discard the rest; uncompressed WAV spends the same amount on everything, which is why it is so large.

How big is a minute of audio in each format?

The figures below assume stereo at 44.1 kHz. A 60 minute podcast recording is therefore about 58 MB as 128 kbps MP3, 84 MB at 192 kbps, 144 MB at 320 kbps, and 606 MB as WAV.

Which bitrate can I actually hear?

Listening tests consistently find that 192 kbps MP3 is transparent for most listeners on most material, meaning they cannot reliably tell it from the uncompressed original. At 128 kbps, cymbals and reverb tails can sound slightly swishy on good headphones, though speech is fine. 320 kbps is the ceiling for MP3 and is worth it only for critical listening or when the file will be edited again, since each lossy re-encode compounds the loss. AAC, the codec inside M4A files and most video, is about 20 to 30 percent more efficient than MP3 at the same bitrate, so a 128 kbps AAC sounds roughly like a 160 to 170 kbps MP3.

What is the difference between MP3, M4A, WAV and FLAC?

MP3 is the lossy format everything plays, from car stereos to thirty-year-old software, which is why it remains the safe choice for sharing. M4A is a container for AAC, Apple's and YouTube's default; it sounds better per kilobit but some older Windows software and cheap players refuse it, hence M4A to MP3. WAV is uncompressed PCM, the format studios record and edit in; 16-bit at 44.1 kHz is CD quality and 48 kHz is the standard for video. FLAC compresses WAV losslessly to about 50 to 60 percent of its size and is the right format for an archive of CD rips or masters.

Converting between lossy formats always costs something. Going from a 128 kbps M4A to a 320 kbps MP3 does not restore detail; it just stores the existing loss in a bigger file. Converting MP3 to WAV likewise produces a large file with the MP3's quality. The only free conversions are lossless to lossless (WAV to FLAC) and copying a stream without decoding it, which is what Extract Audio does with the M4A option.

What sample rate should I use?

44.1 kHz is the CD standard and what music is distributed at; 48 kHz is the standard for anything attached to video, including YouTube, broadcast and film. Kilnfile's MP3 to WAV tool offers both. Match the destination: 48 kHz for a video edit, 44.1 kHz for a music project or a CD. Converting between the two is harmless in practice but pointless if you do not need to. Higher rates such as 96 kHz are for studio production, not for listening, and only make files bigger.

Which setting should I pick for common jobs?

Podcast episode: MP3 at 128 kbps for speech, 192 kbps if there is music, 44.1 kHz. Sharing a song: MP3 at 192 kbps. Voice note or lecture recording: MP3 at 128 kbps or lower, mono if the tool allows. Editing in Audacity, Logic or a video editor: WAV at 48 kHz for video or 44.1 kHz for audio-only. Pulling the soundtrack from a video for a backing track: Extract Audio with M4A, which copies the AAC stream losslessly in a second. Archiving: FLAC, or WAV if space is no object.

Frequently asked questions