Philips Research - Technologies
Philips' Spatial Audio Coding:
|
||
|
A new Philips compression technique, referred to as
Spatial Audio Coding, allows multiple audio channels to
be transmitted to mobile and static audio devices for
only a negligible increase in transmission bit rate
compared to conventional mono or stereo coding. This
effectively means superior audio quality for users
without requiring network operators to take up much more
bandwidth. Importantly, Spatial Audio Coding is also
backwards compatible with existing audio playback
devices. In cooperation with other companies, the
technology is being incorporated in international
application standards such as MPEG. Furthermore, it is
also part of the latest standard for mobile telephony
(3GPP), and set makers have announced to incorporate it
in their mobile phones this year.
Growing demand for audio with more channels There is a continuing trend towards multi-channel audio. Home entertainment systems increasingly support surround sound capabilities, with DVDs or SACDs providing the multi-channel source. There is also a growing demand for better stereo-quality streaming of music to mobile devices, apparently leading to a higher demand for bandwidth. This poses a problem to manufacturers whose main limitation is the storage capacity of mobile devices themselves and to operators whose primary constraint is available bandwidth. Superior audio with minimal bandwidth increase Spatial Audio Coding is an algorithm that addresses all of these issues. By encoding stereo or multi-channel audio signals in a much more efficient way, it allows signals of superior quality (e.g. stereo instead of mono, or multi-channel instead of stereo) to be transmitted for a minimal increase in bandwidth. These signals are stored on a mobile device in their encoded form, and only decoded for playback, which means they take up considerably less memory space than files of similar quality that use other encoding techniques. Spatial Audio Coding operates by extracting the ‘spatial parameters’ of a stereo or multi-channel signal. As an example, it takes a 5.1 signal with five channels (left, right, centre, left surround and right surround) and turns this into a stereo signal (left and right) complemented by an additional signal that contains the spatial data. This spatial signal is extremely small, typically 10% of the stereo signal. True surround sound experience Upon decoding, there are two possible scenarios. If the receiving device is multi-channel enabled, then the spatial data encoded in the transmitted signal will be used to reconstruct a multi-channel version of the stereo signal, providing the listener with a true surround sound experience. The second scenario is that the receiving device cannot process as many channels as have been received (i.e. a stereo device receives a 5.1 signal, or mono device receives a stereo signal). In such situations, the additional spatial information that has been sent as a separate part of the signal is ignored, and the receiving device subsequently produces stereo or mono, as appropriate. Backwards compatibility This backwards compatibility is extremely important because existing audio devices can still receive signals encoded using Spatial Audio Coding without any loss in quality. This means there are two possible ways of benefiting from Spatial Audio Coding: transmitting more channels using virtually the same amount of transmission or storage bandwidth, or transmitting the same amount of channels using considerably less bandwidth. Digital radio broadcasting A prime example of the application of this technology would be in digital radio broadcasting. Transmitting stereo signals typically requires 128kbit/second of bandwidth, whereas transmitting DVD-quality (e.g. 5.1 multi-channel) would require approximately 360kbit/second, an increase of roughly a factor of 3. Using conventional encoding techniques, broadcasters who want to offer multi-channel would therefore be faced with either a significant loss in perceived quality, or an increasing bandwidth usage by almost 300%. Neither is a realistic option. With Spatial Audio Coding, on the other hand, multi-channel sound can be broadcast with only a small increase (typically 10%) in bandwidth. This would allow multi-channel digital radio to be reproduced on surround sound home audio systems. It is also an extremely interesting proposition for in-car audio. Some cars are already equipped with as many as 13 loudspeakers yet only receive stereo radio broadcasts.
For more information:
|
|
