View source for Digital Measurement SEI Audio File

{{Breadcrumb|}} {{Breadcrumb|Digital}} {{Breadcrumb|Digital Streaming Measurement}} {{CurrentBreadcrumb}}
[[Category:Digital]]

The Nielsen SEI server-to-server solution ([[Digital Measurement SEI Content Viewing]]) needs audio files for each watermarked source that requires crediting.

{| class="wikitable"
|-
! '''Audio Compression Type''' !! '''Tested ideal Compression Rate'''
|-
| AC3 Stereo || 192 Kbps or higher
|-
| AC3 5.1 || 384 Kbps or higher
|-
| Enhanced AC3 || 192 Kbps
|-
| MPEG2 audio || 192 Kbps or higher
|-
| AAC with ADTS headers || 
|}

* If source audio is stereo, include both Left and Right channels
* If source audio is 5.1, take L + .707C into Left and R + .707C into Right and deliver stereo
* Audio sampling rate should be 32kHz, 44.1kHz or 48kHz

=== S3 Bucket and Prefix Naming Convention ===
<code>DTVRAudio_<'''partnerid'''>_<'''AssetId'''>_<'''UTCStart'''>_<'''UTCEnd'''></code>
{| class="wikitable"
!'''Name'''
!'''Description'''
|-
|partnerid
|Abbreviation provided by Nielsen for each provider or publisher
|-
|AssetId
|in-house assetId
|-
|UTCStart
|time in 32-bit unsigned int UTC in seconds from first second of audio captured
|-
|UTCEnd
|time in 32-bit unsigned int UTC in seconds from first second of audio captured
|-
|}
*Files should be placed in S3 buckets in a mutually agreed upon folder structure.
*15 minutes of audio does not have to align with hour boundary. For example for broadcast from 1 AM to 2 AM, 4 audio files each with 15 minutes of audio from 1:00 AM to 1:15 AM, 1:15 AM until 1:30 AM, 1:30 AM until 1:45 AM and 1:45 AM until 2:00 AM, but if the encoder turns on at 1:07, deliver through 1:22, etc. The preference is hourly bound but not required.
*If difficult to achieve exactly 15m increments, err towards shorter. I.e. 10m instead of 20m
*Avoid small, fragmented audio files. Watermark decoding algorithms require runway. Too short of a file will result in lost audio codes
*If file times/audio overlap, ingestion will overwrite the portion of the older file
*Precision is important for UTC start/end file naming convention. Use UTC timestamp of Server time/encoder
*The file should not be encrypted.
*One copy per stream only.

=== Audio Asset Metadata ===
== S3 Bucket and Prefix Naming Convention ==
<code>DTVRAssetMetaData_<'''partnerid'''>_<'''AssetId'''>_<'''UTCStart'''>_<'''UTCEnd'''></code>
{| class="wikitable"
!'''Name'''
!'''Description'''
|-
|partnerid
|Abbreviation provided by Nielsen for each provider or publisher
|-
|AssetId
|in-house assetId
|-
|UTCStart
|time in 32-bit unsigned int UTC in seconds from first second of audio captured
|-
|UTCEnd
|time in 32-bit unsigned int UTC in seconds from first second of audio captured
|-
|}

=== SLA ===
The files must be delivered into the proper S3 bucket within 2 hours of the broadcast. For example files from 1:00 AM to 1:15 AM must be delivered before 3:00 AM.