Your browser is out-of-date!

Update your browser to view this website correctly. Update my browser now

×

AWS adds multiview, AI capabilities to Elemental

The new capabilities enable multi-feed viewing, real-time content metadata, and ad monetisation for live streams

Amazon Web Services (AWS) has announced new capabilities for the Elemental portfolio, bringing AI, personalisation and intelligent monetisation to the live video workflow.

AWS Elemental MediaPackage Dynamic Multiview allows viewers to select their own multi-camera layout from a single set of encoded sources, delivered to any device over standard HTTP Live Streaming (HLS) or Dynamic Adaptive Streaming over HTTP (DASH), without per-layout encoding costs.

Stitching multi-camera and multi-feed layouts on demand from encoded sources, MediaPackage Dynamic Multiview delivers a single adaptive bitrate stream through HLS or DASH. Sources encode one time through AWS Elemental MediaLive. Layouts are assembled in the compressed domain with no re-encoding, said the company, with more than 95 per cent of consumer devices reached with a single digit rights management key and no SDK, per-device builds, or new vendors.

MediaPackage Dynamic Multiview stitches multi-camera and multi-feed layouts on demand from encoded sources, delivered as a single adaptive bitrate stream through HLS or DASH. Sources encode once through AWS Elemental MediaLive. AWS Elemental MediaPackage assembles layouts in the compressed domain with no re-encoding, providing cost savings of up to 97 per cent, said AWS. The output reaches more than 95 per cent of consumer devices with a single digital rights management (DRM) key and requires no SDK, per-device builds, or new vendors.

AWS Elemental Inference Contextual Metadata enables personalised video experiences and contextual ad decisioning, extracting AI-generated metadata from live video in real time.

Running in parallel with encoding, computer vision models identify objects, actions, brands, on-screen text, and sentiment from video content. Metadata is emitted via a standard API aligned to the Interactive Advertising Bureau (IAB) content taxonomy and the Global Alliance for Responsible Media (GARM) brand suitability categories, powering downstream workflows such as automated highlight clipping, discovery, contextual ad decisioning through Elemental MediaTailor and content moderation, all without custom ML infrastructure.

AWS Elemental MediaTailor Yield Optimisation automates backfill of unfilled ad inventory using Amazon Publisher Services (APS) demand, providing incremental revenue with zero setup cost.

Available ad slots are filled from Amazon Ads inventory during ad insertion, aiming to monetise dead air and slate video. Incremental revenue is delivered with no enabling cost or additional infrastructure. Ads are inserted server-side, ensuring they play seamlessly across devices, cannot be blocked, and transition without buffering. At launch, MediaTailor Yield Optimisation is available to APS streaming TV publishers. Designed for large-scale live events, it remains brand-safe by filtering demand by category, avoiding competitive conflicts. Publishers set their own price floors and category rules. 

AWS Elemental MediaTailor Low-Latency Ad Insertion uses Apple’s HLS interstitials to extend personalised server-side ad insertion of MediaTailor to low-latency streams at scale. 

Sharing the same design philosophy: intelligence layered inside the video pipeline, not bolted on after, all four capabilities run within the same AWS Elemental services that power millions of streams around the world, offering the same 99.9 SLAs, operation models and no need to onboard new vendors, said the company.