³ÉÈËÂÛ̳

The idea behind making this happen is to treat the individual elements of a radio programme as objects with their own identity and behaviours, much like the way data is represented as objects in object orientated programming. For example in a documentary programme you have a number of different elements, each of which play a different role in the programme. You might classify the different parts of the programme as follows:-

±Ê°ù±ð²õ±ð²Ô³Ù±ð°ù/²Ô²¹°ù°ù²¹³Ù´Ç°ùÌý

* interviews (multiple voices)

* contributors/interviewees (single voices)

* actuality/narrative sound effects (e.g. the sound of a key event)

* background sound effects (giving a sense of place)

* foreground music (illustrating part of the story)

* background music (reflecting the mood or emotion)

* linking music (giving pace and timing or marking a transition)

Currently all these elements are combined on a single timeline in an audio editor and rendered as an audio file, and alternative versions are created by re-editing Ìýthe original version. The listener can turn the volume up and down, jump around in the time line and press pause and play, but cannot affect the mix and the pace of the programme. Some players allow variable speed playback with pitch adjustment, but this makes the voices sound unnatural and harder to understand.Ìý

If, however, the elements of a programme were available as separate objects which interacted with each other in a defined running order then the listener could change the way the parts of the running order interact to make a programme more suited to their requirements. So, if for example my train journey is 25 minutes, but I want to listen to a half hour programme, the player could take all the elements of the programme, knowing which parts are narrative and which parts linking elements and reduce the gaps between the key narrative elements. Alternatively, if I wanted to slow the pace of a programme down to give me more time to think about what was being said, then the player could lengthen the linking sections by playing more of the music or sound effects that join the narrative elements still leaving the speech to run at the same speed that it was recorded.Ìý

Of course if you have a programme available as a series of separate objects then you could also change the relative levels to create a version with reduced levels of music and background sound effects, trading a rich soundscape for improved intelligibility for those who need it. People who hate the use of music in radio programmes could remove it altogether. Some people might even chose to remove the presenter. The player could contain your preferences for radio listening, so it always offered you your preferred listening mode, or it could react to the surrounding conditions as you move around, for example removing background sounds when you are in a noisy environment.Ìý

There have been some explicit examples of things like this in the past, for example allowed you to remix the song 'Digging in the Dirt', but in fact interactive and reactive soundscapes are commonplace as a part of computer gaming. Could we bring more to our audience by delivering object based linear content, as in radio and TV programmes? Could such an approach deliver sufficient improvements for our audience to make a change in the way we make programmes, along the lines of create once but publish for may different individual members of the audience?