Audio Codecs and File Formats

Core Idea

An audio codec (coder-decoder) is an algorithm for encoding audio data, typically to reduce file size, and decoding it for playback. The choice of codec and container format determines audio quality, file size, playback compatibility, and suitability for specific use cases. Professionals must understand the tradeoffs between lossless and lossy codecs to make appropriate format decisions for recording, editing, delivery, and distribution.

Lossless formats preserve every sample exactly as recorded — the decoded audio is bit-for-bit identical to the original. WAV (Waveform Audio File Format) is the dominant uncompressed format in professional audio — a 24-bit/48kHz stereo file stores approximately 17 MB per minute. AIFF (Audio Interchange File Format) is Apple's equivalent, functionally similar. FLAC (Free Lossless Audio Codec) compresses audio without data loss, reducing WAV by approximately 50–60% — useful for archiving and audiophile streaming (Tidal MQA, Amazon Music HD, Apple Music Lossless use FLAC or ALAC). Broadcast WAV (BWF) embeds timecode and metadata into the WAV container, essential for film and broadcast post-production workflows.

Lossy codecs achieve much greater compression (10:1 or more) by perceptual coding: removing audio content the human auditory system is unlikely to notice — frequencies masked by louder simultaneous sounds, content below the hearing threshold, and temporal post-masking. MP3 (MPEG-1 Audio Layer III) was the dominant consumer format for 25 years. AAC (Advanced Audio Coding) is its successor, used by Apple, YouTube, and streaming platforms — at the same bitrate, AAC sounds better than MP3 due to improved psychoacoustic models. Opus is a highly efficient modern codec (used by Spotify, Discord, WhatsApp) that delivers excellent quality at low bitrates (96–160 kbps) through time-frequency adaptive coding.

Bitrate (kbps — kilobits per second) directly controls the quality/size tradeoff in lossy formats. 128 kbps AAC is sufficient for casual listening; 256–320 kbps is transparent to most listeners; 320 kbps MP3 is the highest standard MP3 bitrate. Variable bitrate (VBR) encoding allocates more bits to complex passages and fewer to simple ones, maintaining consistent perceived quality at lower average file size.

Explainer

The proliferation of audio formats reflects the tension between audio fidelity and practical constraints of storage, bandwidth, and computational resources. In the early history of digital audio, storage and bandwidth were genuinely scarce — MP3's ability to compress CD audio by a factor of 10 was transformative, enabling music to move over the early internet and portable digital players.

Today, storage is cheap enough that lossless streaming is commercially viable (Apple Music delivers 256 kbps AAC as its standard, with lossless ALAC to all subscribers at no additional cost). The practical need for lossy compression has shifted from storage to real-time communication — Opus at 48 kbps powers Discord, WebRTC, and voice calls while delivering intelligible audio with acceptably low latency.

For audio professionals, format decisions cascade through workflows. Always work in the highest-resolution lossless format available during production — any lossy encoding should be a deliberate final delivery step, not an intermediate storage decision. Applying effects and processing to already-lossy files compounds the artifacts, particularly with lossy codecs at borderline bitrates. The standard practice is: capture and edit in WAV/AIFF, master in WAV, deliver in the format and bitrate specified by each platform's technical requirements.