Audio content specifications for language versions (dubbed and voice-over)

The audio content of language versions consists of a set of master files to be published on the OTT platform as an alternative soundtrack.

The set of master files includes:

• Audio master file in stereo, bleeped out or no swear words
• Audio master file in stereo, uncensored swear words
• Audio master file in 5.1 surround sound, bleeped out or no swear words
• Audio master file in 5.1 surround sound, uncensored swear words
• Dialogue master file in stereo, uncensored
• Dialogue master file in 5.1 surround sound, uncensored

All changes to this list should be agreed upon in advance with the manager responsible.

All file names must follow the template.

File names can only contain English letters, underscores, and a full stop that separates the file name from the extension. Any special characters or spaces are prohibited.

Language versions should use the original content's name. For information about this name, contact the manager.

File names for series or films cannot be changed during a project.

Any additional information included in the name must be approved by the Customer in advance.

The name template includes a fixed sequence of elements:

• Project name
• Only for TV series: season and episode number in the following format: s00_e00
• Audio format (w/o full stops): 20 (for stereo) or 51 (for 5.1)
• Censorship level (not applicable to dialogue master files): cens—bleeped out or no swear words, uncens—uncensored swear words
• Master file type: DUB—for dubbing, VO—for voice-over, DX—for dialogue master files
• Creation date in the following format: YYYY_MM_DD
• Tag of the film language (only for dialogue master files) as per ISO 639-2
• Special designations, if necessary: e.g., v2, v3, etc.

Examples:

• Container: Wave or Broadcast Wave
• Codec: PCM
• All audio channels are combined into one file
• Channel sequence: L, R
• Sample rate: 48 kHz
• Bit rate: 24 bit
• Duration: all audio master files must have the same duration as the video with the native frame rate (accurate down to milliseconds)
• Integrated loudness (EBU R128): -23 LUFS +/-0.5 LU
• Loudness range (LRA): not more than 18 LU
• Peak signal (True Peak): not exceeding -2 dBTP
• Timecode: if timecode is embedded in BWAV, it must start at 01:00:00:00

• Container: Wave or Broadcast Wave
• Codec: PCM
• All audio channels are combined into one file
• Channel sequence: L, R, C, LFE, Ls, Rs (SMPTE)
• Sample rate: 48 kHz
• Bit rate: 24 bit
• Duration: all audio master files must have the same duration as the video with the native frame rate (accurate down to milliseconds)
• Integrated loudness (EBU R128): -23 LUFS +/-0.5 LU
• Loudness range (LRA): not more than 18 LU
• Peak signal (True Peak): not exceeding -2 dBTP
• Timecode: if timecode is embedded in BWAV, it must start at 01:00:00:00

• Source content:
• PCM soundtracks obtained by decoding compressed formats (both lossy or lossless) are prohibited.
• Compatibility:
• All versions must be synced (if played simultaneously, no chorus, flanger, or tripping effect should occur).
• All versions must include all soundtrack elements. This does not apply to bleeping out swear words in censored versions.
• To avoid issues with playing 5.1 sound on some user devices, automatic downmix from 5.1 to stereo must be controlled using the following settings:
• Center channel attenuation: -3 dB
• Surround channel attenuation: -3 dB
• Downmix does not include LFE
• For full dubbing, take into account sound effects (including spatial) laid over voice in the original version and recreate them as closely as possible to the original.
• Speech synchronization:
• For full dubbing, diegetic speech must be synchronized with the actors' lips as much as possible.
• When creating voice-overs, avoid placing the translation ahead of the original line.
• Speech audibility and sound balance:
• Any speech in the soundtrack must be audible both in terms of actor articulation and sound balance. Effects and music can prevail over speech only if this is the case in the original soundtrack.
• Bleeping out swear words:
• When bleeping out swear words, cover up the entire word root, not just the stressed vowel. Masking sound effects must not make the line inaudible.
• Phasing channels:
• All channels must be phased with each other. If correlometer readings are below zero for a significant length of time, it is considered a defect.
• The soundtrack must not include:
• Micropauses
• Foreign interference absent in the original that prevents the sound from being heard, such as:
• Recording artifacts: hissing, saliva, excessively loud sound of closing/opening lips
• Foreign sounds absent in the original
• Parasitic noises, crackling, clicking
• Electrical interference, background noises
• Audible non-linear distortion
• Distortion caused by triggered limiters and compressors
• Editing defects: truncated lines, clicks at cuts, etc.

Creative solutions that do not meet these requirements are allowed only if similar solutions are used in the original soundtrack.