On Google Ad Manager, publishers can monetize their audio inventory like any other inventory. There are a number of different types of audio inventory available through Ad Manager, from music streaming to live radio and live events, text to speech news articles and smart speaker and display.
Programmatic buying of audio inventory from Ad Manager is supported for third-party buyers through the Authorized Buyer platform via OpenRTB or Authorized Buyers RTB. Auction packages are also available for bidders who are interested in curated inventory for various Audio types.
Jump to: Implementation | playback method values | supported audio environments
Implementation
The following values should be sent to bidders according to the examples for either OpenRTB or Authorized Buyers RTB, and the linked developer documentation.
OpenRTB
OpenRTB defines an Audio object to hold all information relevant to the audio inventory.
audio
{
mimes: "audio/mp3"
mimes: "audio/ogg"
maxduration: 30
protocols: VAST_2_0
protocols: VAST_3_0
"playbackmethod": 5
Authorized Buyers RTB
allowed_video_formats
field of the Video object.allowed_ad_types: ALLOWED_AD_TYPE_AUDIO
allowed_ad_types: ALLOWED_AD_TYPE_VIDEO
is_amp_page: DIALECT_HTML
amp_ad_requirement_type: AMP_AD_NOT_ALLOWED
only_deal_bids_accepted: true
video {
videoad_start_delay: 0
max_ad_duration: 20000
video_ad_skippable: ALLOW_SKIPPABLE
skippable_max_ad_duration: 600000
allowed_video_formats: AUDIO_MP3
allowed_video_formats: AUDIO_OGG
playback_method: INITIATE_ON_ENTERING_VIEWPORT_SOUND_OFF
allowed_video_formats
value designates the file type accepted by the publisher. You should read the playback_method value, which designates the muted state of the inventory.
Buyers should not return audio creatives for requests with SOUND_OFF
values.
Playback method values
METHOD_UNKNOWN = 0
AUTO_PLAY_SOUND_ON = 1
AUTO_PLAY_SOUND_OFF = 2
CLICK_TO_PLAY = 3
MOUSE_OVER = 4
INITIATE_ON_ENTERING_VIEWPORT_SOUND_ON = 5
INITIATE_ON_ENTERING_VIEWPORT_SOUND_OFF = 6
Supported audio environments
Below are a number of supported audio environments that are available to Authorized Buyers:
Streaming playlists
A streaming playlist is the concept of an audio player which plays a set of audio clips specific to an individual user streaming in an online environment. The playlist can be unique to each user, the timing of breaks is scattered across all users, and audio clips are generally short in length (less than 10 minutes), which provides ample opportunity for an ad to be between unique audio clips.
Text-to-speech
Text-to-speech uses machine learning technology to convert written article content into audio formats and deliver audio ads into these opted-in user experiences. Text-to-speech audio content is made available in Ad Manager by Publishers for buyers to transact types just like other audio formats (i.e. podcasts, live radio, smart devices).
Audio in video
Audio demand competes with video demand for âlistenableâ video content (i.e. news, sports, live-stream). Common setups include desktop and mobile implementations that request and render when the video player is in the background. As a best practice, we recommend responding with a companion banner as an overlay on top of the video player. If there is no buyer-provided companion, publishers may default to rendering a default image.
Common examples:
- Music playlists on Spotify, Pandora, and YouTube Music
Streaming podcasts and time-shifted events
A streaming podcast and a time-shifted event are similar to the streaming playlist environment. They also exist in a fully online setting and the timing of the clips are unique to each user. The primary differences are the length and potential for ads already in the audio clip. These differences mean that the decision of whether to show an ad and at what point are more complex, likely involves more mid-rolls, and potentially involves licensing.
Common examples:
- Streaming podcasts in Pandora, iHeart, TuneIn, and Spotify
- Non-live sporting events