CORDIS provides links to public deliverables and publications of HORIZON projects.
Links to deliverables and publications from FP7 projects, as well as links to some specific result types such as dataset and software, are dynamically retrieved from OpenAIRE .
Deliverables
This service offers automatically generated play-lists of diverse forms. The service is available for individual broadcast-stations as well as for a set of stations. Not only radio-, but also music playing TV-stations are observed and reports about their play-lists are generated.
Basis for the service is an automatic fingerprinting technology, which detects and recognises nearly every kind of music automatically. As determined by the technology in use, a music-title has to be introduced into the reporting-system prior to its recognition, and there will be "black spots" in the automatically generated play-lists, where the recognition was not successful. These "black spots" will be reworked by human operators, who will complete the reports manually, thus the customer get a quality-controlled report. Non-music is qualified either as "Non-Music" or as "Commercials", "News" or "Speaker".
The Commercial Monitor is a result build on top of the basic components. It provides a complete, packaged solution for commercial/advertisement -monitoring.
The Commercial Monitor provides an efficient way for detection of commercials in broadcast audio-signals. This can be used either for TV- or radio broadcast. Commercials in the sense of the product are audio-clips with a duration among 5 and 60 seconds.
There are 10 000 commercials, which can be looked for on up to 3 audio-signals in parallel. The systems works offline, the typical configuration provides updates every hour, which guarantees that commercials are detected 2 hours after they have been sent at the latest.
The "Monitor Extension" is not a self-standing independent result. It can only be used in conjunction with one of the previously mentioned "Commercial Monitor" or "Music Monitor" results. Monitor Extension provides additional capacity for monitoring audio-channels. It can be plugged into one of the packaged starter-products. It provides up to 4 additional audio-input channels for the main monitoring product.
By plugging in a Monitor Extension, there is no overall loss of performance for the whole system as long as the total number of monitored channels is below 50. The Monitor Extension simply fits into the other components and can also be managed and controlled from the Operator Work Station, delivered with the Commercial- or Music- Monitor.
The Music Monitor is a result build on top of the basic components. It provides a complete, packaged solution for broadcast-monitoring. From a functional point of view, Music Monitor provides recognition on up to 3 parallel audio-signals, which are fed into the system. The system is designed to compare the input signals with up to 50.000 titles (music, commercials). The minimal title length is 10 seconds, however music should be longer guarantee uniqueness of titles, otherwise the recognition rate might go down. The systems works offline, the typical configuration provides updates every hour, which guarantees that music is detected 2 hours after broadcasting at the latest.
The RAA Basic Components consist of several individual components:
The AudioDNA Extractor is a library implemented in C++. It provides methods for automatic calculation of AudioDNA, which describes the significant features of an audio-signal. Usually each gene represents among 20 and 50ms, the chain of genes is named AudioDNA. Input for this library is an audio-signal with fixed sampling structure. The library is available for Linux as well as on Windows.
Observation Station is a self-standing component, which makes use of the AudioDNA Extractor library and provides infrastructure for the permanent calculation and forwarding of AudioDNA including administration of meta-data. It is available for Linux and Windows.
Acquisition Station is a self-standing component, which makes use of AudioDNA Extractor and external meta-data to provide a descriptor for ingestion into the central database. Thus the station is used to create new (reference) AudioDNA. Available for Linux, soon for Windows.
Content Analysis Component is a self-standing component to compare AudioDNA received from Observation Stations with AudioDNA stored in the reference AudioDNA database. In addition it is the interface for administrative tasks regarding the reference database (e.g. Acquisition Station). Available for Windows and Linux on request.
AudioDNA database is a standard SQL-database, which provides management of the fingerprint-database.
Searching for OpenAIRE data...
There was an error trying to search data from OpenAIRE
No results available