An Open, Standards-Based Framework for Audio Metadata Transport in Live Content Workflows

摘要

Described is a model for open audio metadata transport in live workflows that will meet the requirements for next-generation audio systems such as ATSC 3.0. The framework is standards-based and is essence agnostic. The format groups audio metadata into logical payloads that can be customized for specific use cases. It allows for bit efficient coding of audio metadata, but also allows for full, structured syntaxes, including XML-based syntaxes. The format is also not tied to a specific metadata standard. Current audio metadata standards are supported, but the format is open to support new audio formats as they arise, as well as private and/or nonstandard metadata. Realization of the framework using key-length-value (KLV) for the base structure of the format is proposed. SMPTE ST 2109 targets applications for transport in AES3 streams, for operation with existing serial digital interface (SDI) and AES3-based infrastructures. Methods for transport over internet protocol (IP) are proposed including native KLV transport using Realtime Transport Protocol (RTP) over IP, which can be utilized by ST 2110 media flows. While targeted at live production and distribution, the format takes into consideration metadata paths for emission to enable a seamless “microphone to speaker” metadata path for efficient delivery of audio metadata from content creator to the end user. Interoperation with file-based formats and workflows are also considered in the design.