GB2345233A

GB2345233A - Encoding of multiple digital audio signals into a lesser number of bitstreams, e.g. for surround sound

Info

Publication number: GB2345233A
Application number: GB9823146A
Authority: GB
Inventors: John Robert Emmett
Original assignee: Individual
Current assignee: Individual
Priority date: 1998-10-23
Filing date: 1998-10-23
Publication date: 2000-06-28
Also published as: GB9823146D0

Abstract

A system for encoding multiple digital audio signals, e.g. the front left 1, front right 2, centre 3, rear left 4, rear right 5 and sub-bass channels 6 of a surround sound system, for distribution or recording interleaves the channels so as to give two output bit streams at 13,14. The unfiltered but delayed channels 1 and 2 are combined at 11 to provide one bit stream. The other bit stream is provided by low pass filtering the channels 4 and 5 and combining them at 12 with the delayed channel 3. The channel 6 may be combined, after low pass filtering, with either or both outputs 13,14, and may be carried as allocated user bits in a channel or as least significant bits of a channel. The channel 3 may alternatively be combined with the channels 1 and 2. The bit streams may be decoded by reverse processes.

Description

Apparatus and Method, for the Encoding and Decoding of multiple digital Audio Signals or Bitstreams, into a lesser number of conventional Channels.

This invention relates to a method for the coding and decoding of associated groups of audio signals in digital form.

The encoding and decoding of associated pairs of digital audio signals into a single digital bitstream, is for example, described in the IEC Standard 60958 which describes a protocol known as the"AES/EBU"bitstream. As such, this protocol is in widespread use for the recording and processing of Stereo programme material, both for sound only purposes, or for the sound channels associated with pictures.

Particularly in the field of sound associated with pictures, the use of multiple discrete"Surround Sound"channels is becoming popular, one arrangement being described in the ITU-R Standard BS 775. This Standard describes a group of associated channels consisting of one front centre channel, left and right front channels, left and right rear"surround"channels, and a low bit rate "Sub-Bass"channel.

The coding of these six channels as one combined entity is described for example in US 5632005, and several such methods are in use within Cinema distribution applications. However within the sound production stage, prior to distribution of the finished programme, there would be a considerable advantage if the channels could be maintained in a form whereby each channel was accessible individually or in combination, without affecting other associated channels. One solution to this problem is to arrange storage of the channel groups as associated AES/EBU bitstreams, of which three would be required. However there are only two AES/EBU bitstreams associated with many common Video recorders, and this present method describes one method for the storage and distribution of six associated channels of surround sound in two AES/EBU bitstreams.

Preferred Embodiment Figure 1 shows a block diagram of a system in accordance with the present invention, and Figure 2 shows the basic arrangement of the AES/EBU digital audio protocol, which may be used as a typical carrier medium. The signals are assumed to consist of the six discrete channels described in Recommendation ITU-R, BS 775 or to follow a similar arrangement. The particular arrangement described in Figure 1, consists of ; 1. Front Left, 2.

Front Right, 3. Centre, 4. Rear Left, 5. Rear Right, and 6. Sub-Bass channels.

The basis of the present invention is to combine groups of associated channels into combined bitstreams at 11, forming outputs 13 and 14. Any similar arrangement will suffice to carry the combination signals.

The bit rate reduction of the source signals necessary for this combination to take place, consists of low pass filtering in 8. and 9. The unfiltered channels carry matching signal delays at 7. This process is followed by interleaving of the samples from associated channels in 12. The single channel thus formed has many of the desirable properties of a discrete audio channel, such as the ability to monitor and process signals, yet it can be decoded into two discrete channels by the reverse process to that shown in Figure 1. The order of interleaving is preferably obtained from the block structure pattern present in the AES/EBU protocol that is shown in Figure 2.

Each sub-frame of information (21,22 etc up to 23), consists of a preamble 15, audio and auxiliary data in 16, a validity bit 17, User and Channel status bits 18 and 19, and it ends with a Parity bit at 20. There are a total of 196 sub-frames between 21 and 23, forming a block of data, the start of which is indicated by a unique preamble 15. Where this is absent, or where the block structure has been altered in some way, the least significant bit (s) of the interleaved audio signal itself may be used to signal the interleave structure.

In one embodiment of this invention, the two rear surround channels are beneficially combined into a single channel in 12, It being known art that the rear surround channels require less quality of coding than the front channels.

The very low bitrate Sub-Bass channel may be carried either as allocated User bits within the AES/EBU protocol (18 in Figure 2), or partially interleaved with the least significant bits of one or more channels, or added to the least significant bits.

Similarly, the Centre (front) channel is conventionally used to carry only Dialogue, which requires less coding quality than is required for the Left and Right front channels. Therefore, the Centre channel could be partially interleaved with the Left and Right front channels, the degree of interleaving being adjusted to the relative quality requirements of the LeftIRight and Centre channels. The resulting combination channel could be decoded as a conventional Stereo source, with the dialogue content being equally distributed between Left and Right channels.

Other possible arrangements depend on the available recording or distribution bitrate. For instance, given the capability to carry a full 24 bits in each audio sample, the allocation of the least significant bits to the rear surround information would enable the carriage of this information in a perceptually hidden form within the bitstream carrying the conventional Stereo channels.

Claims ; 1. A system for the encoding of two or more associated digital audio signals, whereby conventional digital Pulse Code Modulated samples representing the audio channels are interleaved in a known manner in order to form a single data stream, suitable for recording or distribution using conventional digital apparatus.

2. A system for decoding two or more associated digital audio signals, whereby the digital Pulse Code Modulated samples retrieved from conventional recording or distribution apparatus, are de-interleaved in a known manner, forming two or more discrete data streams. The data streams thus formed, are suitable for reproduction, processing, or onwards recording or distribution using conventional apparatus.

3. A method of obtaining the interleave pattern of the encoded samples from the received data pattern.

4. A method of obtaining the interleave pattern of the encoded samples from the received audio data.

5. A method of incorporating an interleaved channel between one or more other channels, such that the proportion of interleaved samples can be adjusted according to the fixed or instantaneous quality demands placed on each channel.

6. A method whereby auxiliary data such as Metadata is incorporated as part of the interleaving process.

Claims

Amendments to the claims have been filed as follows Claims; 1. A method for the coding of an associated pair, or associated pairs of Pulse Code Modulated digital audio signals, which themselves form elements of a larger group of associated surround sound signals, whereby: (a) the samples representing the pair of audio channels to be coded are low pass filtered and interleaved in a known manner in order to form a single Pulse Code Modulated data stream; and (b) the resultant coded Pulse Code Modulated data stream is reincorporated as part of the larger group of associated surround sound signals.
2. The coding method of claim 1, further comprising a method for coding an additional low bit rate audio channel or channels, by a process of interleaving the channel data in a known manner with the least significant bit, or bits, of the coded samples.
3. A method whereby the interleave pattern of the samples coded according to claim1, or coded according to claim 1 and claim 2, may be determined from the received audio data samples alone.
4. A method of coding an interleaved channel according to claiml or claim 2, such that the relative proportion of the interleaved samples may be adjusted adaptively according to the fixed or instantaneous quality demands placed on each channel.
5. A method whereby auxiliary data such as Metadata is incorporated as part of the interleaving process according to claim2.