AD Export
Closed Caption Creator offers three different export workflows for audio description projects. The broadcast export produces finished audio deliverables — voiceover tracks and a mixed-down program file — suitable for direct delivery to a broadcaster or streaming platform. The script file export produces timed text documents intended for use by live voice talent in a recording session or post-production pipeline. The extended AD export is a specialized workflow that produces a restructured video package where AD narration is inserted into pauses added to the program video itself, as required for certain accessibility standards.
Broadcast Export
To begin a broadcast export, confirm that the correct AD Event Group is selected, then click the export icon to open the File Export window. Choose Audio Description from the list of options and click Next to open the Audio Description Exporter.
Export Folder. Choose a destination folder for the output files. The exporter generates a separate voiceover audio file for each AD event, a combined VO mixdown that merges all events into a single continuous track without program audio, and a program mixdown that contains the program audio and voiceover together as a final composite deliverable.
Event Group. The Event Group dropdown is filtered to show only Audio Description type groups. Select the group you want to export. By default the application pre-selects the group that is currently active in the editor.
Export Profile. Choose the audio format for the output files. Supported formats are FLAC, MP3, and WAV. Most broadcasters require a Broadcast WAV file with a 48 kHz sample rate, and this is the recommended selection for broadcast delivery.
Export Selection. You can export the entire Event Group or limit the export to a time selection. Exporting a selection is useful when you need to regenerate only a portion of the deliverable — for example, when pickups have been recorded for a subset of events.
Mix Preset. The Mix Preset determines how the program audio is ducked whenever a voiceover event plays. Presets define the fade-in, fade-out, gain reduction depth, and timing of the ducking applied to the program audio. Presets labeled "high ratio" apply more aggressive ducking and are better suited to content with loud background music or sound effects where the standard preset would allow the program audio to compete with the voiceover.
Loudness Profile. The Loudness Profile applies EBU R128 or ATSC A/85 normalization to the output. You can choose to apply loudness processing to the voiceover only, the program audio only, or the final mixdown. Because most broadcasters deliver program audio that has already been mastered to a loudness standard, the recommended practice is to apply loudness normalization to the voiceover track only. Modifying the program audio is typically outside the scope of the AD work unless the client has specifically requested it.
When all settings are confirmed, click Export Audio. The application re-renders any events that have changed since the last render pass, assembles all audio files with the correct timing offsets, applies the mix and loudness processing, and writes the output files to the selected folder.
BWF Metadata (WAV Only)
When exporting to Broadcast WAV format, the exporter provides an optional BWF Metadata section that embeds Broadcast Wave Format chunk data directly into the WAV file header. This metadata travels with the file and is readable by DAWs, broadcast playout systems, and audio quality-control tools without requiring a separate sidecar file.
The BWF Metadata section is disabled by default. Enable it by checking Enable BWF Metadata within the Advanced Options area of the exporter. Three fields are available.
Title stores the name of the program or deliverable. It defaults to the project name and can be edited to match whatever identifier the receiving facility expects. This field maps to the Description field of the BWF bext chunk, which many systems expose as the file title in their asset management panels.
Description provides a longer free-text note that can be used for production context, version information, or client instructions. It is also written into the bext chunk and is surfaced by applications that read extended WAV metadata.
Reference Time specifies the absolute timecode position at which the audio file should be placed when the WAV is imported into a DAW or playout system. It defaults to the project's program in-code — so if the project starts at 01:00:00:00 the exported WAV will carry that timecode as its reference point. This allows the receiving engineer to drop the file onto a timeline and have it snap to the correct position automatically, without needing to read a separate cue sheet or spot the audio manually. The value is stored as a sample offset within the bext chunk and is calculated from the entered timecode at the project's frame rate.
If the receiving facility has not specified a requirement for BWF metadata, leaving this section disabled has no effect on audio quality or compatibility — the resulting WAV files are standard PCM audio and will play back normally in any application that accepts WAV format.
Some broadcast workflows require an audio description control track as defined in the BBC White Paper WHP198. A control track is a two-channel WAV file where Channel 1 contains the mono AD narration and Channel 2 carries encoded metadata — fade and pan information modulated as an audio signal — that instructs downstream broadcast-mix or receiver-mix systems to duck the program audio automatically whenever narration plays.
Closed Caption Creator supports this workflow through an integration with Engine Desktop by Emotion Systems. Export your AD voiceover and timing data from Closed Caption Creator, then import those files into Engine Desktop to generate a broadcast-ready WAV file containing both the narration and the encoded control track.
For a full walkthrough of this workflow, see the blog article: Audio Description with Control Track.
Script File Export
When the project requires live voice talent, the AD script needs to be exported in a format that the recording team and the talent can work from. To export a script file, open the File Export window, choose Subtitle File from the options (not Audio Description), and click Next.
The SubRip (.srt) format is one of the most common choices for AD script delivery. It is a plain-text format that contains timecodes and description text and is widely understood by post-production tools and studios.
For teams that prefer word processing formats, the LMS Transcript profile for Microsoft Word includes timecode, description text, and the notes field from each event. This is useful when the notes field has been used for production comments or director instructions that need to accompany the script.
For quality control workflows, the Excel export with the DVW QC Script or DVW Studio Script profile is a strong option. The QC Script profile includes timecode, description text, speaker IDs, event notes, replies, and other metadata in a structured spreadsheet that a QC reviewer can annotate directly.
If your organization uses an internal database or custom tooling, the CSV and JSON profiles provide raw structured data that can be ingested programmatically. If a profile format your team needs is not available, contact the support team to request a custom profile.
Extended AD Export
Standard audio description works within the natural silences of the program audio. Extended audio description is a different approach in which the video itself is paused to allow longer narration windows, and then resumes playback after the description has been delivered. This is required for content where the dialog or action is so dense that adequate narration cannot fit within the existing silences.
To enable the extended AD workflow, open the Audio Description Exporter as described in the Broadcast Export section above, and check the Export as Extended AD option. In addition to the standard audio output, the exporter will also generate an SRT subtitle file containing the timecodes and description text for all events. This SRT file is used in the next stage of the workflow to reconstruct the video.
The video reconstruction is completed in a non-linear editing application such as Adobe Premiere Pro. Import the exported SRT file and the audio files into your project. Each subtitle marker in the SRT represents a point where the video must be paused to insert a description. At each marker position, cut the video, move the downstream segment to align with the end of the inserted audio, and create a frame hold over the gap — a freeze of the last frame before the cut that plays while the AD audio is delivered. This process is repeated at every marker throughout the timeline until all description windows have been inserted and the reconstructed video is ready for final export.
The extended AD mixdown produced by Closed Caption Creator includes the voiceover audio at the correct offsets for the restructured timeline. Loudness processing on the voiceover track is still applicable; the program audio normalization considerations that apply to standard AD export apply here as well.