Background

Over the last ten years, live broadcast audio has moved beyond the established standard of 5.1 surround sound. In 2019, BT Sport (now TNT Sports) launched their English Premier League UHD service with Dolby Atmos audio. As UHD services rolled out worldwide, the associated refresh of silicon created an opportunity to deploy a new range of audio codecs with new capabilities. Broadcast standards such as ATSC 3.0 helped drive the development of new features in these Next Generation Audio (NGA) codecs. The Dolby NGA emission codec for delivery to consumer devices is Dolby AC-4.

While a better immersive experience is a key part of NGA, personalization is also important. Customers now expect to be able to select their language of choice or select dialogue level. The hard of hearing and partially sighted expect accessibility features such as Dialogue Enhancement and Audio Description. The flexibility of audio delivery systems like Dolby AC-4 enables new features to be incorporated without requiring new silicon in consumer devices. However, to bring these features to life, more audio channels are required upstream of the audio emission encoder, as well as a metadata audio description. 

NGA use-cases typically require more audio channels than legacy use-cases like 5.1 surround. For example, live immersive audio typically uses a 5.1.4 channel configuration thus requiring 10 channels instead of the 6 for 5.1. To provide personalization options, audio elements must be separated when fed to the NGA audio encoder. This requires more channels than before where only a single complete main mix plus a stereo downmix was required. Remote production (REMI), where raw microphone feeds are brought back to the Network Operations Center (NOC) requires higher channel counts than traditional models.

As well as the extra channels, personalization requires a description of the audio elements being presented to the encoder. This is now possible with the standardization (ITU-R BS 2125) and adoption of Serialized ADM. The SMPTE ST 2110 suite of standards solved the problem of increase audio channel requirements and carriage of Serialized ADM for uncompressed audio workflows.

While many compressed contribution and distribution links have transitioned to IP over the last few years, the need for audio compression has not disappeared. While primary links often use IP with uncompressed audio, satellite is still a popular option for back-up circuits to redundancy using a completely different circuit path. Cloud ingress or egress and intercontinental links are still sufficiently expensive to make compressed audio attractive. As the industry transitions to REMI and NGA, a new compressed audio solution is required. As well as providing more channels and metadata capability, such a solution should address other industry concerns such as lower latency and higher compression efficiency.

Introducing Dolby IIF Pro

Dolby's proposal for compressed contribution and distribution of NGA is Dolby IIF Pro, an implementation of MPEG AAC-IIF. AAC-IIF is a new audio standard currently being drafted by MPEG with publication expected mid 2026. While being developed with in-home and car-projection use-cases in mind, the feature-set of AAC-IIF was identified to be a perfect match for contribution/distribution use-cases. The codec is based upon proven AAC technology but provides advantages not available to existing members of the MPEG-4 audio family.

Key Capabilities

• Scalable channel count of up to 256, all phase aligned (without profile-related restrictions)
• Audio Transparency achieved at 128kbps per mono channel
• End to end latency as low as 12.5ms
• Optional carriage of serialized ADM metadata as per ITU-R BS.2125
• Video frame alignment

Dolby IIF Pro addresses all known NGA use-cases as well as future and Dolby legacy use-cases by carrying metadata via Serialized ADM. ADM is a versatile extensible system which can adapt as requirements for distribution evolve.

Another key advantage is guaranteed phase alignment across all audio channels. Existing solutions typically support large channel counts with multiple stereo instances. This approach risks comb filtering effects on the audio due to small misalignments between audio streams.

Through standardization and implementation efforts, Dolby IIF Pro is expected to be compatible with all commonly used transports for contribution and distribution. As a member of the MPEG 4 audio family, existing support for MPEG 4 audio in containers such as MPEG-2 TS or MPEG-DASH, and associated transports such as SRT or RIST can be leveraged. Future work includes standardizing MPEG AAC-IIF in SMPTE ST 337 to enable carriage by SDI or SMPTE ST 2110-31.

Use Cases

• Venue to NOC contribution or REMI
• International primary distribution
• Ground to cloud or cloud to ground

Key advantages of Dolby IIF Pro include scalability, bit-rate efficiency and low latency making it ideal for live contribution and distribution applications. It provides the right mix of features and flexibility to deliver the future of live audio.

Dolby has a long history providing solutions designed to help deliver the excitement of live events. We believe Dolby IIF Pro is the right solution to continue that tradition.

aac-iif-white-paper.pdf