Music content ingestion and publishing

Non-stop global content ingestion & publishing

A substantial part of Omnifone’s MusicStation platform is the constantly running, massively scalable, behind the scenes global content ingestion service. Omnifone’s catalogue is one of the largest in the world providing access to more than 25 million tracks. MusicStation powers the content behind more services in more countries than any other platform.

Based on enterprise grade infrastructure and non-stop architecture, this automated global ingestion hub manages the complex content lifecycle from label delivery through service availability and beyond.

The end-to-end process includes ingestion from thousands of global label sources, super-aggregation, matching, de-duplication, cleansing, rights digestion,asset manufacture, DSP analysis, DRM preparation, asset marshalling and publishing to hundreds of territory services.

Omnifone is the champion of super high-quality global catalogue provisioning in an industry notorious for its haphazard approaches. Our commitment to quality is backed up by constant investment in automated and human de-duplication and cleansing. Partnering with the best third-party sources, content metadata is enhanced and decorated to an unrivalled quality.

Delivering to the biggest names behind today’s global digital music services, our catalogues are subject to the most stringent SLAs for quality, availability and particularly frontline content punctuality. Similarly our audio engineering and asset manufacturing capabilities are second to none – because our sound engineers are second to none.

High quality global, regional & local catalogue

With a huge database of more than 25 million songs, MusicStation provides instant access to the entirety of the world’s musical collection in a single location. Serving more services in more territories than anyone else, Omnifone is committed to maintaining and constantly growing the best, biggest and most relevant catalogue.

With three ingestion centres covering the globe, Omnifone constantly ingests content from all the major international and regional sources and thousands of locally relevant labels. We understand better than anyone else that musical taste varies everywhere.

Omnifone’s localisation and regional relevancy processes include multi-language support for application display and search capabilities. We guide our content acquisition based on local label intelligence and detailed profiling of music consumed in target countries. This is also backed up by automated local chart and radio play comparisons.

Omnifone works tirelessly in the support of international standards for content ingestion, delivery and reporting. We are board members of DDEX and founder members of the Global Repertoire Database (GRD).

Massively parallel content factory

The global ingestion hub constantly pulls in content from thousands of sources via hundreds of feeds. This massively scalable enterprise infrastructure is capable of ingesting, preparing and publishing over 100,000 tracks per day. Priority based ingestion automatically promotes tracks based on release date and popularity - a P1 will take no longer than one hour from receipt to service publication, including both human and automated quality monitoring.

Ingestion feeds of all types are supported using standards such as DDEX or the proprietary formats of labels and aggregators. New feeds can be setup extremely rapidly and DDEX feeds can be configured the same day. Labels who do not support digital delivery can instead benefit from our upload tools with integrated Gracenote metadata matching. Our flexible, highly-efficient ingestion tools & processes can support usage specific multi-service provisioning from a single source feed.

Source media is ingested in lossless HD audio and transcoded to multiple device formats via a massively parallel processor farm. This hugely scalable job infrastructure is also used for encryption and DRM file & license preparation. In addition many digital signal processing technologies are run across the catalogue extracting recognition fingerprints, acoustic metadata and determining mood characteristics.

A powerful set of tools support the content lifecycle from ingestion through to publishing. This includes: catalogue management, de-duplication, content expedition, parental advisory, language localisation, metadata enhancement, image management, content rights & publishing, takedown control and end-to-end visibility via an audit trail.

Formats for every device and network

MusicStation supports a huge range of audio formats, bit rates, and container structures to maximise device and network support. Audio CODEC formats include HE-AAC v1, HE-AAC v2, AAC-LC, MP3, WMA and FLAC. Supported bit rates are 24, 48, 64, 192 & 320 kbps and container packaging supports the 3GPP, MP4, OMA, Marlin, ADTS and HMPP formats.

Omnifone treats audio quality with the highest importance - we are known internationally for our talent and expertise. As a Dolby Media Generator partner, MusicStation utilises Dolby Pulse, a highly optimised codec that provides the highest audio efficiency and music quality for all manner of devices. It is particularly well suited to online and mobile audio environments, ensuring faster download times and reliable media streaming in situations where bandwidth fluctuates.

No other platform comes close to this breadth of cross-device audio support, flexibility and quality.

Rich enhanced metadata bringing apps to life

MusicStation collates vast amounts of information from multiple sources and musicologist compiled libraries, about each and every song. Enhanced music metadata describes and defines music in a detailed way to help deliver the best global music features possible.

Data is acquired from record label feeds, publishers, DSP analysis, musicologist curation and it is significantly enhanced using third party services such as Gracenote and Rovi.

Enhanced metadata includes a sophisticated three-tier genre system, which goes deep down into a micro-genre level, artist era and a DSP extracted two-tier mood hierarchy. Audio acoustic data is also DSP extracted providing tempo, average BPM, intro, outro and hook points.

Substantial metadata is available at the artist level including artist type (e.g. solo female, boy band), artist origin (e.g. USA,

MX, JP), multi-lingual biographies, profiles and reviews and multiple size and quality artist images. High resolution front & back cover album artwork is also available and dynamically resizable. There are also a wealth of explorable relationships between recording performers, contributors and composers.

Additional rich music metadata includes feeds from chart companies and popularity & ratings sourced from internal listening stats and third-parties. Tracks and albums also carry parental advisory explicit flags. MusicStation comes with a sophisticated classical dataset with classical composers and support for compositions, works, meta-works and an editorially assigned ‘works of importance’.

All of this rich metadata is easily available via MusicStation Warp and Mars ensuring that customers can tune applications to deliver exactly the experience that users are looking for. Cross referencing this wealth of metadata against a user’s listening and skip habits over time can drive very powerful profiling and recommendations.