Learn
Audio File Formats Explained
Audio files come in more formats than most people realise, and the differences between them are more than cosmetic. The format you choose determines how large your files are, how much audio detail they retain, and how widely they can be played back. Whether you are recording a podcast, mastering a record, or simply trying to keep your music library in order, understanding the landscape of audio formats saves time and prevents mistakes that are difficult to undo.
This guide walks through the major categories of audio encoding, explains each common format in practical terms, and helps you decide which one fits your work.
The three categories: uncompressed, lossless, and lossy
Every audio format falls into one of three buckets, and the distinction matters more than any individual codec name.
Uncompressed audio stores the raw digital representation of a sound wave with nothing removed or altered. The data that comes off an analogue-to-digital converter is written directly to disk. This means perfect fidelity but very large files. WAV and AIFF are the two dominant uncompressed formats.
Lossless compression reduces file size without discarding any audio data. A lossless file can be decoded back to a bit-perfect copy of the original uncompressed audio, much like a ZIP archive recreates its contents exactly. FLAC and ALAC are the main lossless codecs, and they typically compress audio to around 50 to 60 per cent of the equivalent uncompressed size.
Lossy compression achieves much smaller files by permanently removing audio information that the codec's psychoacoustic model deems inaudible or less important. Once that data is gone, it cannot be recovered. MP3, AAC, and OGG Vorbis are all lossy formats. At high bitrates, the difference from the original may be imperceptible to most listeners, but repeated re-encoding of lossy files degrades quality with each generation.
For anyone managing a growing collection of audio, understanding these categories is the first step toward a well-organised cloud library that does not waste space on redundant copies or lose quality unnecessarily.
The major formats
WAV
WAV (Waveform Audio File Format) is the workhorse of professional audio. Developed by Microsoft and IBM, it stores uncompressed PCM (pulse-code modulation) audio and is compatible with virtually every piece of software and hardware in existence. A standard CD-quality WAV file uses a sample rate of 44.1 kHz and a bit depth of 16 bits, though the format supports higher specifications as well.
The trade-off is size. A three-minute stereo track at CD quality runs to roughly 30 MB. At 24-bit/48 kHz, the standard for professional recording, that figure climbs higher still. WAV is the default choice for recording, editing, and mixing in a DAW, and it remains the safest format for archival when lossless compression is not an option.
AIFF
AIFF (Audio Interchange File Format) is Apple's counterpart to WAV. It stores the same uncompressed PCM audio and produces files of the same size. For practical purposes, WAV and AIFF are interchangeable in terms of quality. AIFF is more commonly encountered in Mac-based studios and in Logic Pro workflows, while WAV dominates everywhere else. If you are collaborating across platforms, WAV is the safer bet for compatibility.
FLAC
FLAC (Free Lossless Audio Codec) compresses audio without losing a single bit of information. A typical FLAC file is 50 to 60 per cent the size of the equivalent WAV, and when decoded, it reconstructs the original audio exactly. FLAC is open-source, widely supported on Android, Windows, and most dedicated audio players, and has become the standard for archival and high-quality distribution outside the Apple ecosystem.
For music creators who want to archive sessions or deliver high-quality masters without the storage penalty of uncompressed audio, FLAC is an excellent choice. It is also the format of choice for many digital music stores that sell lossless downloads.
ALAC (Apple Lossless)
Apple Lossless Audio Codec does the same job as FLAC but within Apple's ecosystem. It offers similar compression ratios and identical audio quality to FLAC. The key difference is compatibility: ALAC works natively in iTunes, Apple Music, and all Apple devices, while FLAC requires third-party support on iOS and macOS. If your workflow is centred on Apple hardware, ALAC may be the more convenient lossless option.
MP3
MP3 (MPEG-1 Audio Layer III) is the format that made digital music distribution possible. Released in the early 1990s, its lossy compression algorithm can shrink an audio file to roughly a tenth of its uncompressed size while retaining acceptable quality for casual listening. At 320 kbps (the highest standard MP3 bitrate), most listeners cannot distinguish an MP3 from the lossless source in a blind test.
MP3's universal compatibility is its greatest strength. Every device, every operating system, every media player supports it. For podcasters, MP3 remains the default distribution format because of this universality, and understanding bitrate choices helps you balance file size against listening quality.
At lower bitrates (128 kbps and below), MP3 artefacts become more noticeable: a slight swirling in the high frequencies, a loss of stereo width, and a thinning of transients. These compromises are the cost of small file sizes.
AAC
AAC (Advanced Audio Coding) was designed as the successor to MP3, and by most technical measures it succeeds. At the same bitrate, AAC produces better-sounding audio than MP3, with fewer artefacts and more natural high-frequency reproduction. It is the default codec for Apple Music, YouTube, and most modern streaming platforms.
AAC at 256 kbps (Apple's standard for purchased music) is widely considered transparent, meaning indistinguishable from lossless for the vast majority of listeners. For content creators distributing audio online, AAC offers a good balance between quality and file size, though MP3 may still be preferred when maximum compatibility is the priority.
OGG Vorbis
OGG Vorbis is an open-source lossy codec that competes directly with MP3 and AAC. It is patent-free and generally delivers quality comparable to AAC at equivalent bitrates. You encounter Vorbis most often in Spotify's desktop streaming (at 320 kbps for premium subscribers) and in video game audio, where its open licensing makes it attractive to developers.
For most end users, OGG Vorbis is not a format you choose to encode into so much as one you encounter through platforms that have chosen it for you. Its open-source nature and solid performance make it a respectable codec, but its narrower playback support compared to MP3 and AAC limits its usefulness as a delivery format.
Opus
Opus is the newest major lossy codec and arguably the most technically impressive. Developed by the Internet Engineering Task Force, it excels at low bitrates, making it ideal for voice calls, live streaming, and real-time communication. At higher bitrates, it competes with AAC and Vorbis. Opus powers the audio in WhatsApp calls, Discord, and many WebRTC applications.
For voice-heavy content like AI voice notes and voice memos, Opus is hard to beat: it produces clear, intelligible speech at bitrates where MP3 would sound muddy and artificial.
File size in practice
To make the differences concrete, here is what a typical three-minute stereo song looks like across formats at standard settings.
As uncompressed WAV at 16-bit/44.1 kHz, the file weighs roughly 30 MB. Encoded as FLAC, that drops to approximately 15 MB with no loss of quality. A 320 kbps MP3 comes in around 7 MB, while a 128 kbps MP3 shrinks to about 3.5 MB. AAC at 256 kbps sits at roughly 5 to 6 MB.
These numbers scale predictably. A ten-track album in WAV might occupy 300 MB; the same album in FLAC takes around 150 to 180 MB; in high-quality MP3, roughly 70 MB. When you are managing hundreds or thousands of files, these differences compound. A clear understanding of how much cloud storage you need depends partly on which formats you choose to keep.
Choosing by use case
Recording and production
Record in WAV (or AIFF if you prefer). Use 24-bit depth and 48 kHz sample rate as your default. This gives you headroom for editing and mixing, and WAV's universal DAW support avoids compatibility surprises. Never record directly to a lossy format; you want the full audio data available for processing. Keeping your audio project files well organised from the start means you will not lose track of which files are originals and which are compressed copies.
Archival and backup
FLAC is the natural choice for long-term storage. It preserves every bit of the original audio at roughly half the file size of WAV. If your archive lives within the Apple ecosystem, ALAC achieves the same goal. Either way, archiving in a lossless format means you can always generate lossy versions later without going back to the original session. A reliable backup strategy should include your lossless masters as a priority.
Music distribution and streaming
AAC and MP3 dominate here. AAC at 256 kbps is the standard for digital music purchases and streams well. MP3 at 320 kbps remains the universal fallback when you need maximum device compatibility. Most streaming platforms transcode your upload to their preferred format anyway, so uploading a lossless master and letting the platform handle conversion is the best approach.
Podcasts
MP3 is still the standard for podcast distribution. Nearly every podcast directory and player supports it without issue, and the spoken-word content of most podcasts does not demand the fidelity that music does. A mono MP3 at 128 kbps is perfectly adequate for speech. AAC is technically superior, and Apple Podcasts supports it well, but MP3's universality makes it the safer choice for reaching every listener.
Voice memos and quick captures
AAC or Opus, depending on your device. Modern smartphones typically record voice memos in AAC by default, which is sensible: the files are small, the quality is more than sufficient for speech, and transcription tools handle AAC without difficulty. Opus is increasingly used in apps that prioritise minimal file size for voice.
When format matters and when it does not
Format selection matters most at the extremes of the workflow: at the point of recording (where you want maximum quality) and at the point of delivery (where you want the right balance of quality, size, and compatibility). In between, during editing and mixing, you are typically working with whatever format your DAW uses internally.
For casual listening, the differences between a 320 kbps MP3 and a FLAC file are, for most people on most playback systems, negligible. The format debate is most consequential for archival, professional delivery, and situations where files will be re-encoded or processed further. If you plan to apply audio search or AI analysis to your files, starting with higher-quality formats gives those tools more data to work with.
It is also worth noting that converting between formats only moves in one direction gracefully. You can always create an MP3 from a WAV, but you cannot meaningfully recover the discarded data by converting an MP3 back to WAV. The resulting WAV will be larger but will not sound any better. This is why keeping lossless masters matters, even if you distribute in lossy formats.
Frequently asked questions
What is the difference between lossy and lossless audio?
Lossy compression permanently removes audio data to achieve smaller files. The discarded data cannot be recovered. Lossless compression reduces file size without removing any data; the original audio can be perfectly reconstructed from the compressed file. Uncompressed formats like WAV store raw audio data with no compression at all.
Is FLAC better than MP3?
FLAC preserves the full quality of the original audio, while MP3 discards some data to achieve smaller files. For archival, editing, or critical listening, FLAC is the better choice. For casual listening on portable devices or streaming, a high-bitrate MP3 (320 kbps) is often indistinguishable from FLAC to most listeners on typical playback equipment.
Can I convert MP3 to FLAC to improve quality?
No. Converting a lossy file to a lossless format increases the file size but does not restore the audio data that was discarded during the original lossy encoding. The resulting FLAC will sound identical to the MP3 source. Always keep your original lossless or uncompressed files if quality matters.
What bitrate should I use for MP3?
For music, 320 kbps offers the best quality MP3 can provide and is suitable for most listening. For speech and podcasts, 128 kbps mono is generally sufficient. Bitrates below 128 kbps introduce noticeable artefacts in music. See the audio bitrate guide for a more detailed breakdown.
Why do DAWs use WAV instead of FLAC?
WAV's uncompressed nature means there is no decoding overhead; the DAW can read and write audio data directly without processing it through a codec. This matters when handling dozens or hundreds of tracks simultaneously. Some modern DAWs do support FLAC for import and export, but WAV remains the standard for active sessions because of its simplicity and speed.
Is AAC better than MP3?
At the same bitrate, AAC generally produces higher-quality audio than MP3, with fewer artefacts and better handling of complex audio. AAC at 256 kbps is roughly equivalent to MP3 at 320 kbps in perceived quality. However, MP3 has slightly broader device compatibility, particularly with older hardware.
What format does Spotify use?
Spotify uses OGG Vorbis for its desktop and mobile apps, streaming at up to 320 kbps for premium subscribers and 160 kbps for free-tier users. Spotify's web player uses AAC. The platform transcodes all uploads to its own formats regardless of what the artist submits.
Should I keep WAV files or convert everything to FLAC?
For active projects, keep your WAV files in their original format to avoid any compatibility issues with your DAW. For archival, FLAC saves significant storage space (roughly 40 to 50 per cent) with zero quality loss. Many producers maintain WAV for current work and FLAC for completed projects, which is a sensible compromise.
What is the best format for sharing audio files?
It depends on the recipient and purpose. For professional collaboration, WAV or FLAC ensures no quality loss. For clients who need to listen but not edit, a high-quality MP3 or AAC file is convenient. Sharing large audio files through a cloud workspace avoids the compression that email attachments impose and keeps everything accessible.
Does audio format affect how well AI can transcribe it?
Higher-quality formats give transcription algorithms more audio data to work with, which can improve accuracy, particularly in noisy recordings or with multiple speakers. That said, modern transcription services handle common lossy formats like MP3 and AAC well. The recording conditions (microphone quality, background noise, speaker clarity) typically matter more than the file format.