The bridge for personalisation

Development of an audio-analysing system with the focus on
facilitating segmentation of audio files. It will consist of
modules i.E. Speaker recognition system, speech analysis
systems and audio indexing.

Re-using audio content broadcast by radio stations is not as easy as it might seem. The difficulty lies in not having content ready segmented (e.g. speech from music). The Bridge for Personalisation Project's aim is to facilitate the segmentation of existing and future audio content in order for it to be used in personalised web-services or have it archived for later reuse. During recent years MP2, MP3 and RealAudio have become everyday tools for radio stations for two reasons. Firstly, radio programme exchange and streaming audio via the Internet have enhanced the competitiveness of radio stations because web radio is a cheap distribution channel for audio content. Secondly, in 1998, it became possible to gain near broadcast quality with a low bandwidth of 56kBits/s (standard modem) due to improvement of audio codecs. In order to facilitate the segmentation of radio output, analogue-linear audio content has to be converted into digital segments. After this process it can be used for digital archiving, indexing, personalised audio streaming etc. Keywords: audio, analysis, segmentation.
Project ID: 
2 626
Start date: 
Project Duration: 
Project costs: 
860 000.00€
Technological Area: 
Speech Processing/Technology
Market Area: 

Raising the productivity and competitiveness of European businesses through technology. Boosting national economies on the international market, and strengthening the basis for sustainable prosperity and employment.