Audio for Broadcasting

Two dominant standards govern broadcast loudness. EBU R128 (Europe, UK, Australia, parts of Asia, most of the rest of the world) specifies integrated loudness of -23 LUFS, true-peak ceiling of -1 dBTP, and a Loudness Range (LRA) recommendation of 7-15 LU for typical content. ATSC A/85 (North America) uses -24 LKFS — equivalent to -24 LUFS, the K- vs L-prefix is regional notation only — with the same -1 dBTP ceiling. Both standards measure 'integrated' loudness across the full programme, with gating that ignores silences below -70 LUFS and short-term values 10 LU below the integrated target. Streaming platforms diverge: -16 LUFS Apple Podcasts, -14 LUFS Spotify and YouTube, -14 LUFS Netflix. Always confirm the delivery spec — broadcasters reject files that miss target by more than ±1 LU.

BWF (EBU 3285) is a strict superset of WAV. The audio payload is identical (PCM in a RIFF container), but BWF adds a mandatory 'bext' (broadcast audio extension) chunk containing: Description (256 chars), Originator (32 chars), OriginatorReference (32 chars), OriginationDate (10 chars, YYYY-MM-DD), OriginationTime (8 chars, HH:MM:SS), TimeReferenceLow/High (sample-accurate timecode start), Version, UMID (Unique Material Identifier), and CodingHistory. The timecode field is the critical element — it allows pre-edited audio to be aligned automatically against a video timecode in MCR or playout systems. Pro Tools, Reaper, Nuendo, Audition all read and write bext correctly; Audacity needs the BWF Metadata extension. Verify before delivery with the BBC's open-source bwf-meta-edit tool.

48 kHz / 24-bit linear PCM is the universal broadcast standard. 48 kHz aligns with digital video clocks (NTSC and PAL both phase-lock to 48 kHz on professional gear), avoiding the resampling that occurs if 44.1 kHz audio is laid against 48 kHz video. 24-bit gives 144 dB of representable dynamic range — comfortable for the wide loudness range of broadcast content (drama, sport, music) and headroom for the loudness-and-true-peak processing chain. Some commercial radio specs accept 16-bit deliverables; television is uniformly 24-bit. Higher rates (96 kHz) are uncommon in broadcast workflows because they double bandwidth without audible benefit on broadcast playout chains. Record at the project rate from the start; sample rate conversion at delivery time is a common error vector and is detectable by quality control.

Lossy codecs (AC-3, AAC, MP2) used in broadcast distribution can introduce inter-sample peaks 0.5-1 dB above the digital sample peaks of the source file. A WAV mastered to 0 dBFS digital can exceed 0 dBFS true-peak after Dolby Digital encoding for ATSC distribution, causing audible clipping on consumer receivers. EBU R128 and ATSC A/85 therefore specify a -1 dBTP ceiling measured with a true-peak meter (4x oversampling). Use a true-peak limiter on the master bus: FabFilter Pro-L 2, NuGen ISL 2, iZotope Ozone Maximizer (true-peak mode), or free TDR Limiter 6 GE. Set ceiling to -1 dBTP, lookahead 5-10 ms, release 100-300 ms. See [what-is-audio-limiter](/guide/what-is-audio-limiter) for limiter operation.

AES31 (Audio Engineering Society Standard 31, parts 1-4) governs the interchange of timecoded audio metadata between systems. AES31-2 defines the disk format (wrapping BWFs in a structured FAT32-friendly directory). AES31-3 defines the ADL (Audio Decision List) format — XML-like text that describes edits, cross-fades, and routing in a DAW-neutral way. Pro Tools and Nuendo can both export AES31 ADLs; broadcasters use them to round-trip an edit between facilities without losing the edit decisions. UMID (Unique Material Identifier) per ISO/IEC 4037 is embedded in the BWF bext chunk and uniquely identifies each piece of audio across the broadcast facility. Confirm the broadcaster's exact AES31 expectations during the delivery contract — many large broadcasters (BBC, NPR, ARD) publish full delivery spec sheets online.

Linear TV and radio playout: BWF (PCM 48 kHz / 24-bit) with timecode and bext metadata. Lossy distribution to consumer platforms: AC-3 (Dolby Digital, 192-448 kbps) or E-AC-3 (Dolby Digital Plus, 192-768 kbps) for ATSC television; AAC (HE-AAC v2 32 kbps for low-bandwidth, AAC-LC 128-192 kbps for primary streams) for OTT and IP radio; MP2 (Layer II, 192-384 kbps) for legacy DVB and some EBU radio links. Studio-to-transmitter links (STL): AES3 digital audio over coax/AES50; or IP codecs (Opus 64-128 kbps, AAC-LD, APT-X Enhanced) for low-latency contribution. Studio-to-studio file delivery: FTP/SFTP of BWF or LiveWire/Ravenna/AES67 for live IP audio. Convert source masters via [WAV to MP3](/convert/wav-to-mp3) or [WAV to AAC](/convert/wav-to-aac) when a lossy proxy is required for review or quick distribution.

Every facility runs incoming files through automated QC before air. Standard checks: integrated LUFS within ±1 LU of target, true-peak below -1 dBTP, sample rate exactly 48 kHz, bit depth 24-bit, channel layout matching contract (mono / stereo / 5.1), timecode start matching the slate, BWF bext fields present and correctly populated, no inter-sample peaks above 0 dBTP, no DC offset, and silence margins at file head and tail per spec. Tools: NuGen MasterCheck Pro for loudness compliance, EBU PLOUD orchestra meters, BBC Audio Tool Kit (BATS) for BWF validation, iZotope Insight 2 for full broadcast-spec metering. Fail any check and the file bounces back to you. Run the same toolchain at your end before delivery — fixing a -22.4 LUFS file to -23.0 LUFS at home costs minutes; rejection and resubmission costs hours and credibility.