TW201528251A - Apparatus and method for efficient object metadata coding - Google Patents

Abstract

An apparatus (100) for generating one or more audio channels is provided. The apparatus (100) comprises a metadata decoder (110) for receiving one or more compressed metadata signals. Each of the one or more compressed metadata signals comprises a plurality of first metadata samples. The first metadata samples of each of the one or more compressed metadata signals indicate information associated with an audio object signal of one or more audio object signals. The metadata decoder (110) is configured to generate one or more reconstructed metadata signals, so that each of the one or more reconstructed metadata signals comprises the first metadata samples of one of the one or more compressed metadata signals and further comprises a plurality of second metadata samples. Moreover, the metadata decoder (110) is configured to generate each of the second metadata samples of each reconstructed metadata signal of the one or more reconstructed metadata signals depending on at least two of the first metadata samples of said reconstructed metadata signal. Moreover, the apparatus (100) comprises an audio channel generator (120) for generating the one or more audio channels depending on the one or more audio object signals and depending on the one or more reconstructed metadata signals. Furthermore, an apparatus for generating encoded audio information comprising one or more encoded audio signals and one or more compressed metadata signals is provided.

Description

High-efficiency object metadata coding device and method thereof

The present invention relates to sound source encoding/decoding, and more particularly to spatial sound source encoding and spatial sound source object encoding, and more particularly to high efficiency object metadata encoding.

Spatial sound source coding tools are well known in the art, for example, there are standardized specifications in the surround MPEG standard. The spatial sound source encoding starts from the original input channel, for example, five or seven channels identified according to its position in the reproduction scheme, namely, left channel, middle channel, right channel, left surround channel, right surround Channel and low frequency enhancement channels. Spatial sound source encoders typically derive at least one downmix channel from the original channel, and additionally derive parameter data about spatial cues, such as inter-channel level differences in channel coherence values, inter-channel phase differences, sound Differences in time between roads, etc. At least one downmix channel is transmitted to the spatial source decoder along with the parametric auxiliary information indicating the spatial cues. The spatial sound source decoder decodes the downmix channel and associated parameter data, and finally obtains an output channel that is approximately the same version as the original input channel. The setting of the channel at the output is usually fixed, for example, 5.1 channel format or 7.1 channel format.

Such channel-based audio source formats are widely used to store or transmit multi-channel audio content, and each channel has a specific speaker at a given location. A faithful reproduction of these types of formats requires a speaker device whose speaker is placed in the same position as the speaker used during the production of the audio signal. Increasing the number of speakers can improve the real three-dimensional virtual reality sound field, but it is increasingly difficult to perform this requirement, especially in a home environment like a living room.

An object-based approach can be used to overcome the need for special speaker devices in which the speaker signal is specifically translated for the playback scheme.

For example, spatial source object encoding tools are well known in the art and are The standard has been established in the MPEG SAOC standard. Starting from the original channel, spatial source code encoding begins with a source object that is not automatically designed for a particular translation rendering scheme, as compared to spatial source encoding. In addition, the position of the sound source object in the reproduction scene is changeable and can be determined by the user by inputting specific translation information to the spatial sound source object codec. In addition, the translation information, that is, the location information to be placed in the specific sound source object in the reproduction scheme, is transmitted by additional auxiliary information or metadata. To obtain a particular data compression, the number of source objects is encoded by a SAOC encoder that drops the mixture based on the particular downmix information to calculate at least one transport channel from the input object. In addition, the SAOC encoder calculates parameterized side information that represents inter-object cues, such as object level differences (OLD), object coherence values, and the like. In spatial sound source coding (SAC), inter-object parameter data is calculated for individual time tiles/frequency tiles, ie, for specific frames of the source signal, for example, 1024 or 2048 samples, 24, 32 or 64, etc., consider the frequency band such that there is parameter data for each frame and for each band. As an example, when a source chip has 20 frames and when each frame is subdivided into 32 bands, the number of time/frequency tiles is 640.

In an object-based approach, the sound field is described as a separate source object. This requires object metadata that describes the time varying position of each source in 3D space.

In the prior art, the first data encoding and coding concept is the spatial sound description exchange format (SpatDIF), and the audio scene description format is currently under development [1]. The audio scene description format is an object-based sound scene exchange format that does not provide any means of compressing object tracks. SpatDIF uses the text-based Open Sound Control (OSC) format for the structure of object metadata [2]. However, a simple text-based performance is not an option for compressed transmission of object trajectories.

In the prior art, another metadata concept is the Sound Source Scene Description Format (ASDF) [3], which has the same disadvantages as a text-based solution. This data is constructed by an extension of Synchronized Multimedia Integration Language (SMIL), a subset of Extensible Markup Language (XML) [4, 5].

Another metadata technique in the prior art is the audio binary format of the scene (AudioBIFS), part of the binary format MPEG-4 standard [6, 7], which is highly relevant to the Virtual Reality Modeling Language (VRML), which has Developed for virtual 3D scenes and virtual reality [8]. The complex AudioBIFS standard uses scene graphs to specify the path the object moves. The main disadvantage of AudioBIFS is that it is not designed for real-time operating systems, which delays the real-time operating system and requires random reading of the data stream. In addition, the encoding of the object location does not utilize the limited viewer's ability to locate. When the listener in the virtual scene of the source has a fixed position, the object data can be quantized to a lower bit value [9]. Therefore, the encoding of object metadata applied to AudioBIFS is not valid for data compression.

If you can improve the encoding of the object metadata, you can get a high degree of appreciation.

The object of the invention is a technique for improving the encoding of object metadata. The object of the present invention is achieved by the apparatus of claim 1, the apparatus of claim 8, the system of claim 14, the method of claim 15, the method of claim 16, and the computer program of claim 17.

The present invention provides an apparatus for generating at least one source channel, wherein the apparatus includes a metadata decompressor for receiving at least one compressed metadata signal. Each compressed metadata signal includes a plurality of first metadata samples. The first metadata sample value in each compressed metadata signal is indicative of information associated with at least one source object signal. The metadata decoder is configured to generate at least one re-established metadata signal, such that each reconstructed metadata signal includes a plurality of first metadata samples of at least one of the compressed metadata signals and further includes a plurality of second elements Data sample value. The metadata decoder generates each of the second metadata samples of each of the reconstructed metadata signals in accordance with at least two of the plurality of first metadata samples of the reconstructed metadata signal. In addition, the apparatus includes a source channel generator that generates at least one source channel in accordance with at least one source object signal and at least one reconstructed metadata signal.

In addition, the present invention provides an apparatus for generating encoded source information, the encoded source information including at least one encoded sound source signal and at least one compressed metadata signal. The apparatus includes: a metadata decoder for receiving at least one original metadata signal, wherein each of the original metadata signals includes a plurality of metadata samples, wherein each of the original metadata signals is sampled in the metadata The value indicates the source signal of the signal with at least one source object signal Corresponding information, wherein the metadata encoder is configured to generate at least one compressed metadata signal such that each compressed metadata signal includes at least one of a metadata sample value of one of the original metadata signals a first group, and any metadata sample value such that the compressed metadata signal does not include at least one of the other at least two of the plurality of metadata samples of one of the plurality of original metadata signals. In addition, the device includes a sound source encoder for encoding at least one sound source object signal to obtain at least one encoded sound source signal.

Moreover, the present invention provides a system for generating information for encoding source information, the encoded source information including at least one encoded sound source signal and at least one compressed metadata signal. Moreover, the system includes means for receiving at least one encoded sound source signal and at least one compressed metadata signal, and generating at least one sound source channel in accordance with the at least one encoded sound source signal and in accordance with the at least one compressed metadata signal.

According to an embodiment, a data compression concept for object metadata is provided that is a valid compression architecture that achieves a limited data rate transmission channel. In addition, a good compression ratio for pure azimuth changes is achieved, such as camera rotation. In addition, this architecture supports discontinuous trajectories, such as positional hopping. In addition, low decoding complexity can be achieved. In addition, random reads at limited reinitialization times can be achieved.

Moreover, the present invention provides a method for generating at least one source channel, wherein the method includes receiving at least one compressed metadata signal, wherein each compressed metadata signal includes a plurality of first metadata samples a value, wherein the first metadata sample value in each compressed metadata signal indicates information associated with at least one source object signal; generating at least one reconstructed metadata signal such that each reconstructed metadata signal includes at least one Compressing a plurality of first metadata sample values of one of the metadata signals, and further comprising a plurality of second metadata sample values, wherein the step of generating the at least one reconstructed metadata signal comprises a plurality of steps according to the reconstructed metadata signal At least two of the metadata samples to generate each of the second metadata samples of each of the reconstructed metadata signals; generating at least one source channel according to the at least one source object signal and according to the at least one reconstructed metadata signal .

In addition, the present invention further discloses a method for generating encoded source information, The code source information includes at least one encoded sound source signal and at least one compressed metadata signal. The method includes: receiving at least one original metadata signal, wherein each original metadata signal includes a plurality of metadata sample values, wherein a metadata sample value of each original metadata signal is indicative of at least one audio source object The information associated with the signal source signal of the signal.

Generating at least one compressed metadata signal such that each compressed metadata signal includes a first group of at least two of a plurality of metadata samples of one of the original metadata signals, and such that the compressed metadata signal is not Any metadata sample value of the second group of the other at least two of the plurality of metadata samples comprising one of the plurality of original metadata signals.

Encoding at least one source object signal to obtain at least one encoded sound source signal.

Furthermore, the present invention provides a computer program that implements the above method when the computer program is executed on a computer or a signal processor.

100‧‧‧Devices for generating (multiple) source channels

100‧‧‧ device

101‧‧‧Source input data

110‧‧‧ metadata decoder

120‧‧‧Source channel generator, sound source decoder

200‧‧‧mixers, blocks

200‧‧‧Selective pre-translator/mixer, pre-translator/mixer, mixer

210‧‧‧ metadata encoder

220‧‧‧Source encoder

250‧‧‧Devices and devices for encoding source information

300/500‧‧‧USAC encoder

300‧‧‧core encoder

400‧‧‧ origin, metadata compressor

400‧‧‧ metadata compressor, block

400‧‧‧OAM encoder

410‧‧‧Location of the source object, precise location of the source object

415‧‧‧ Straight line

420‧‧OAM decoder

500‧‧‧Output interface, origin

501‧‧‧Source output data, data

510‧‧‧Location of the first source object, location, position of the first source object, speaker

511, 512, 513, 514‧‧‧ speakers

520‧‧‧ Position, position, speaker of the second source object

600‧‧‧ mode controller

611‧‧‧Quantification

612‧‧‧Sampling

621‧‧‧Anti-up sampling

622‧‧‧linear interpolation

640‧‧‧Polygon approximation

721‧‧‧Anti-up sampling

722‧‧‧Linear interpolation, linear interpolation

740‧‧‧Steps, Reverse Polygon Conversion

800‧‧‧Selective SAOC encoder, SAOC encoder

900‧‧‧Connect

1100‧‧‧ input interface

1200‧‧‧object processor, processor

1205‧‧‧ Output channel

1210‧‧‧Object Translator

1220‧‧‧ Mixer

1300‧‧‧core decoder, USAC decoder

1400‧‧‧ metadata decompressor

1400‧‧OAM decoder

1600‧‧‧ mode controller

1700‧‧‧post processor

1710‧‧‧ binary translator, binary translator

1720‧‧‧Format conversion, format converter

1727‧‧‧fast

1730‧‧‧32ch speaker, output, output interface

1800‧‧‧Selective SAOC decoder, SAOC decoder, block

1810‧‧‧ Vector reference amplitude shift level, VBAP

1 is a diagram of an apparatus for generating at least one source channel in accordance with an embodiment.

2 is a diagram of an apparatus for generating encoded sound source information according to an embodiment, the encoded sound source information includes at least one encoded sound source signal and at least one compressed metadata signal.

Figure 3 is a diagram showing a system in accordance with an embodiment.

Figure 4 is a diagram showing a portion of a sound source object represented by an azimuth, elevation, and radius from an origin in a three-dimensional space.

Figure 5 is a diagram showing the portion of the source device assumed by the source channel generator and the speaker scheme.

Figure 6 illustrates metadata encoding in accordance with an embodiment.

Figure 7 illustrates metadata decoding in accordance with an embodiment.

Figure 8 illustrates metadata encoding in accordance with another embodiment.

Figure 9 is a diagram showing metadata decoding in accordance with another embodiment.

Figure 10 is a diagram showing metadata encoding in accordance with another embodiment.

Figure 11 is a diagram showing metadata decoding in accordance with another embodiment.

Figure 12 is a diagram showing a first embodiment of a 3D sound source encoder.

Figure 13 is a diagram showing a first embodiment of a 3D sound source decoder.

Figure 14 is a diagram showing a second embodiment of a 3D sound source encoder.

Figure 15 is a diagram showing a second embodiment of a 3D sound source decoder.

Figure 16 is a diagram showing a third embodiment of a 3D sound source encoder.

Figure 17 is a diagram showing a third embodiment of a 3D sound source decoder.

FIG. 2 illustrates an apparatus 250 for generating encoded sound source information, the encoded sound source information including at least one encoded sound source signal and at least one compressed metadata signal, according to an embodiment.

The device 250 includes a metadata encoder 210 for receiving at least one original metadata signal. Each original metadata signal contains a plurality of metadata samples. The metadata sample value of each of the at least one original metadata signal is indicative of information associated with the source object signal of at least one source object signal. The metadata encoder 210 is configured to generate at least one compressed metadata signal such that each compressed metadata signal can include a first group of at least two of the metadata samples of one of the original metadata signals, and The compressed metadata signal does not include any metadata sample values of at least one of the other at least two of the plurality of metadata samples of one of the plurality of original metadata signals.

In addition, the device 250 includes a sound source encoder 220 for encoding at least one sound source object signal to obtain at least one encoded sound source signal. For example, the source channel generator can include an SAOC encoder that encodes at least one source object signal in accordance with conventional techniques to obtain at least one SAOC transport channel and as at least one encoded source signal. Various other encoding techniques for encoding at least one source object channel may alternatively or additionally encode the at least one source object channel.

1 is a diagram of an apparatus for generating at least one source channel in accordance with an embodiment.

Apparatus 100 includes a metadata decoder 110 for receiving at least one compressed metadata signal. Each compressed metadata signal includes a plurality of first metadata samples. The first metadata sample value of each compressed metadata signal is indicative of at least one source Information related to the signal. The metadata decoder 110 is configured to generate at least one re-established metadata signal, such that each reconstructed metadata signal includes a plurality of first metadata samples of at least one of the compressed metadata signals and further includes a plurality of second Metadata sampled value. In addition, the metadata decoder 110 generates each second metadata sample value of each reconstructed metadata signal according to at least two of the plurality of first metadata samples of the reconstructed metadata signal.

In addition, the device 100 includes a sound source channel generator 120 that generates at least one sound source channel according to at least one sound source object signal and at least one reconstructed metadata signal.

When referring to metadata sampling, it should be noted that metadata sampling is characterized by its metadata sample values and the time points associated therewith. For example, such a point in time may be related to the beginning of a sequence of sound sources or its analogs. For example, the symbol n or k can identify the location of the metadata sample within the metadata signal and thereby indicate the (correlated) time point (which is related to the start time). It should be noted that when two metadata samples are correlated with different time points, the two metadata samples are different from other metadata samples, even when their metadata samples are the same. Case.

The above embodiments are based on the discovery that the metadata information associated with the source object signal (included in the metadata signal) often changes slowly.

For example, the metadata signal can indicate location information on the source object (eg, azimuth, elevation, or radius to define the location of the source object). It can be assumed that the position of the source object does not change or only slowly changes most of the time.

Alternatively, the metadata signal may, for example, indicate the volume (eg, gain) of the source object, and may also assume that the volume of the source object changes slowly over most of the time.

For this reason, there is no need to pass (complete) metadata information at each point in time. Conversely, (complete) metadata information is only periodically transmitted at specific points in time, such as periodically, such as at the Nth time point, at time points 0, N, 2N, 3N, and the like. On the decoder side, metadata at intermediate points in time (eg, time points 1, 2...N-1) can then be approximated based on metadata samples of at least two points in time. For example, metadata samples at time points 1, 2...N-1 can be approximated based on time samples 0 and N metadata samples, such as linear interpolation. As mentioned earlier, such approximations are based on the fact that the metadata information found on the source objects typically changes slowly.

For example, in various embodiments, three metadata signals specify the location of a source object in 3D space. The first of the metadata signals can, for example, specify the azimuth of the location of the source object. The second of the metadata signals can, for example, specify the elevation angle at which the source object is located. The second of the metadata signals may, for example, specify a radius associated with the distance of the source object.

Referring to Figure 4, as shown, the azimuth, elevation, and radius clearly define the location of a source object in 3D space.

Figure 4 is a diagram showing the position 410 of the source object represented by the azimuth, elevation and radius from the origin 400 in a three-dimensional (3D) space.

Examples of elevation angles, such as specifying the line from the origin to the position of the object and the angle between the orthogonal projections of the line on the xy plane (the x-axis and the plane defined by the y-axis). The azimuth, for example, defines the angle between the x-axis and the orthogonal projection. By specifying the azimuth and elevation, a line 415 from the origin 400 and the position 410 of the source object can be defined. By further specifying the radius, the precise position 410 of the source object can be defined.

In an embodiment, the azimuth angle is defined as: -180° < azimuth 180°, the elevation angle is defined as: -90° Elevation angle 90°, the unit of the radius can be defined, for example, as a meter [m] (greater than 0 m or equal to 0 m).

In another embodiment, it can be assumed that all x values of the source object position in the xyz coordinate system are greater than or equal to zero, and the range of azimuth can be defined as -90°. Azimuth 90°, the range of elevation angle can be defined as: -90° Elevation angle 90°, the unit of the radius can be defined, for example, as a meter [m].

In another embodiment, the metadata signal can be adjusted such that the range of azimuth angles is defined as: -128° < azimuth The range of 128° and elevation angle is defined as: -32° Elevation angle 32° and radius can be defined, for example, as a logarithmic scale. In some embodiments, the original metadata signal, the compressed metadata signal, and the reconstructed metadata signal may include a zoomed representation of the location information and/or a scaled representation of the volume of one of the at least one source object signals.

The source channel generator 120 can be used, for example, to generate at least one source channel in accordance with at least one source object signal and in accordance with the reconstructed metadata signal, wherein the reconstructed metadata signal can, for example, indicate the location of the plurality of source objects.

Figure 5 depicts the portion of the source of the sound source assumed by the source channel generator and Speaker solution. The origin 500 of the xyz coordinate system is shown in the figure. Additionally, the location 510 of the first source object and the location 520 of the second source object are depicted in the figures. In addition, FIG. 5 illustrates a scheme in which the sound source channel generator 120 generates four sound source channels of four speakers. The sound source channel generator 120 assumes the positions of the four speakers 511, 512, 513, and 514 in FIG.

In FIG. 5, the position 510 at which the first source object is located is close to the assumed speakers 511 and 512 and away from the speakers 513 and 514. Therefore, the sound source channel generator 120 can generate four sound source channels such that the first sound source object 510 is played by the speakers 511 and 512 without being driven by the speakers 513 and 514.

In another embodiment, the sound source channel generator 120 can generate four sound source channels such that the first sound source object 510 is played at a high volume by the speakers 511 and 512 and played at a low volume by the speakers 513 and 514.

In addition, the position of the second source object is close to the assumed position of the speakers 513 and 514 and away from the speaker 511. Therefore, the sound source generator can generate four sound source channels such that the second sound source object 520 is played by the speakers 513 and 514 without being driven by the speakers 513 and 514.

In another embodiment, the sound source channel generator 120 can generate four sound source channels such that the second sound source object 520 is played at a high volume by the speakers 513 and 514 and played at a low volume by the speakers 511 and 512.

In an alternate embodiment, only two metadata signals are used to specify the location of the source object. For example, when all sound source objects are assumed to be in a single plane, for example only orientation and radius can be specified.

In another embodiment, only a single metadata signal is encoded and transmitted as location information for each source object. For example, only the azimuth can be specified as the positional information of the source object (eg, it can be assumed that all of the source objects are in the same plane with the same distance from the center point, and thus are assumed to be the same radius). The azimuth information can be used, for example, to determine that the location of the source object is similar to the left speaker and away from the right speaker. In this case, the source channel generator may, for example, generate at least one source channel to cause the source object to be played by the left speaker and not by the right speaker.

For example, Vector Base Amplitude Panning (VBAP) can be used to determine the width of the source object signal in each source channel of the speaker (see, for example, Ref. [12]). For example, with respect to VBAP, it is assumed that the source object is associated with a virtual sound source.

In various embodiments, another metadata signal may specify the volume of each source object, such as a gain (eg, expressed in decibels [dB]).

For example, in FIG. 5, the first gain value may be specified by another metadata signal of the first source object at location 510, and the second gain value is by another metadata of the second source object at location 520. The signal is specified, wherein the first gain value is greater than the second gain value. In this case, the volume of the first source object played by the speakers 510 and 520 is greater than the volume of the second source object played by the speakers 513 and 514.

Various embodiments also assume that such gains of the sound source object typically change slowly. Therefore, there is no need to transmit such metadata information at each point in time. Conversely, metadata information is only transmitted at specific points in time. At intermediate points in time, the metadata information can be approximated and transmitted, for example, using the metadata samples described above and subsequent metadata samples. For example, linear interpolation can be used for approximation of intermediate values.

In this way, the efficiency of metadata transmission can be effectively saved.

Figure 3 is a diagram showing a system in accordance with an embodiment.

The system includes a device 250 for generating encoded sound source information, the encoded sound source information including at least one encoded sound source signal and at least one compressed metadata signal.

In addition, the system includes a device 100 for receiving at least one encoded sound source signal and at least one compressed metadata signal, and generating at least one sound source channel according to the at least one encoded sound source signal and according to the at least one compressed metadata signal.

For example, when the apparatus 250 for encoding using the SAOC encoder is used to encode at least one source object, at least one encoded source signal can be decoded by the apparatus 100 for generating at least one source channel, the at least one source channel being The prior art uses a SAOC decoder to obtain at least one source object signal.

The location of the object is considered only as an example of metadata to allow random reading at limited reinitialization times, while in various embodiments there is a location to provide periodic retransmission of all objects.

According to an embodiment, the device 100 receives random access information, wherein for each compressed metadata signal, the random access information refers to accessing the signal portion by one of the compressed metadata signals, wherein at least one other of the metadata signals The signal part is not marked by random access information. And the metadata decoder 110 is responsive to the first metadata sample value of the access signal portion of the compressed metadata signal, but not to any other first metadata sample value of any other signal portion of the compressed metadata signal. To generate at least one of the reconstructed metadata signals. In other words, by specifying random access information, a portion of each compressed signal can be designated, while other portions of the compressed signal are not specified. In this case, only a specific portion of the signal is compressed and no other portion is reconstructed as one of the reconstructed metadata signals. Therefore, it is possible to rebuild for a specific point in time, and the plurality of first metadata samples transmitted as the compressed metadata signal represent the complete metadata information of the compressed metadata signal (however, for other time points, the metadata information is not Will be transmitted).

Figure 6 illustrates metadata encoding in accordance with an embodiment. According to various embodiments, the metadata encoder 210 can be used to implement the metadata encoding illustrated in FIG.

In Figure 6, s(n) may represent one of the original metadata signals. For example, s(n) may, for example, represent a function of the azimuth of one of the source objects, and n may indicate time (eg, by indicating a sampling location within the original metadata signal).

The time-varying trajectory element s(n) is sampled at a sampling rate significantly lower than the source sampling rate (eg, equal to or lower than 1:1024) and quantized by a factor of N (see 611) and downsampling (See 612). This produces a periodic transmission digital signal labeled z(k).

z(k) is one of at least one compressed metadata signal. E.g, Each Nth metadata sample of (n) is also a metadata sample of the compressed metadata signal z(k), between each Nth metadata sample. The other N-1 metadata samples of (n) are not sampled for the metadata of the compressed metadata signal z(k).

For example, assume that within s(n), n indicates time (eg, by indicating the location of the sample within the original metadata signal), where n is a positive integer or zero. (eg start time: n=0). N is the downsampling factor. For example, N = 32 or any other suitable downsampling factor.

For example, the downsampling at 612 is used to derive the compressed metadata signal z from the original metadata signal, which can be understood, for example, as follows: z(k)= (k.N); where k is a positive integer or 0 (k=0, 1, 2,...)

Therefore z(0)= (0); z(1)= (32); z(2)= (64); z(3)= (96),...

Figure 7 illustrates metadata decoding in accordance with an embodiment. The metadata decoder 110 in various embodiments may be used to implement the metadata decoding illustrated in FIG.

According to the embodiment illustrated in FIG. 7, the metadata decoder 110 is configured to generate at least one compressed metadata signal by upsampling, wherein the metadata decoder 110 is configured according to the reconstructed metadata. At least two of the plurality of first metadata samples of the signal are linearly interpolated to produce each second metadata sample value for each reconstructed metadata signal.

Thus, each reconstructed metadata signal contains all metadata samples of its compressed metadata signal (this sample is referred to as "a plurality of first metadata sample values" of at least one compressed metadata signal).

Additional ("second") metadata samples are added to the reconstructed metadata signal by performing upsampling. The upsampling step is used to determine the location of the additional ("second") metadata sample values that are added to the reconstructed metadata signal.

By performing linear interpolation, a metadata sample value of a plurality of second metadata sample values is determined. Linear interpolation is performed based on two metadata samples of the compressed metadata signal (the compressed metadata signal has become a plurality of first metadata samples of the reconstructed metadata signal).

According to an embodiment, sampling is performed by performing linear interpolation and sampling values of the second metadata value are generated, for example, in a single step.

In Figure 7, the inverse upsampling process (see 721) combined with linear interpolation (see 722) results in a rough approximation of the original signal to the original signal. The inverse upsampling process (see 721) and the linear interpolation process (see 722) can be performed in a single step.

For example, the inverse upsampling process (see 721) and linear interpolation (see 722) on the decoder side can be performed as follows: s'(k.N) = z(k); where k is a positive integer or 0

Where j is a positive integer and is defined by the following formula: 1 j N-1.

Here, z(k) is a metadata sample that reliably receives the compressed metadata signal z. The value, z(k-1), is the metadata sample value of the compressed metadata signal z, and z(k-1) is received immediately before the compressed metadata signal z(k) is surely received.

Figure 8 illustrates metadata encoding in accordance with another embodiment. According to an embodiment, the metadata encoder 210 can be used to implement the metadata encoding illustrated in FIG.

In an embodiment, as depicted in FIG. 8, in metadata encoding, a good structure can be specified by the difference in encoding between the delay compensated input signal and the linear interpolation coarse approximation.

According to such an embodiment, the inverse upsampling process in conjunction with linear interpolation is also performed as part of the metadata encoding on the encoder side (see 621 and 622 in Figure 6). In addition, the inverse upsampling process (see 621) and linear interpolation (see 622) can be performed in a single step.

As described above, the metadata encoder 210 is configured to generate at least one compressed metadata signal such that each compressed metadata signal includes at least two of the metadata samples of one of the original metadata signals. The first group, the compressed metadata signal can be considered to be associated with the original metadata signal.

Each metadata sample value, which is included in one of the at least one original metadata signal and included in the compressed metadata signal and associated with the original metadata signal, can be considered as a plurality of first One of the metadata samples.

In addition, each metadata sample value, which is included in one of the at least one original metadata signal but not included in the compressed metadata signal and associated with the original metadata signal, is a plurality of second One of the metadata samples.

According to the embodiment of FIG. 8, the metadata encoder 210 performs linear interpolation in accordance with at least two of the plurality of first metadata samples of one of the at least one original metadata signal for the plurality of Each of the plurality of second metadata sample values of one of the original metadata signals produces an approximate metadata sample value.

In addition, in the embodiment of FIG. 8, the metadata encoder 210 is configured to generate a difference for each of the second metadata samples of one of the at least one original metadata signals, such that the interpolation represents the second The difference between the metadata sampling and the approximate metadata sampling of the second metadata sample.

In the preferred embodiment described in FIG. 10, for at least one original At least one of a plurality of differences of the plurality of second metadata samples of one of the metadata signals to determine whether each difference is greater than a threshold.

In the embodiment of Fig. 8, the approximate metadata sample value can be determined, for example, by performing upsampling and linear interpolation on the compressed metadata signal z(k). Upsampling and linear interpolation can be performed on a portion of the metadata encoded on the encoder side (see 621 and 622 in Figure 6), and the same method can be found in the metadata decoding of 721 and 722.

s"(k.N)=z(k); where k is a positive integer or 0

Where j is an integer and: 1 j N-1.

For example, in the embodiment illustrated in FIG. 8, when metadata encoding is performed, a plurality of differences may be determined for differences within 630.

s(n)-s"(n), for example, all n values of (k-1).N<n<k.N, or for example, (k-1).N<n k. All n values of N.

In various embodiments, at least one difference of such is transmitted to the metadata decoder.

Figure 9 is a diagram showing metadata decoding in accordance with another embodiment. According to various embodiments, the metadata decoder 110 can be used to implement the metadata decoding illustrated in FIG.

As described above, each reconstructed metadata signal includes a plurality of first metadata samples of at least one of the compressed metadata signals, the reconstructed metadata signal being considered to be associated with the compressed metadata signal.

In the embodiment of FIG. 9, the metadata decoder 110 generates a plurality of second metadata samples in each reconstructed metadata signal by generating a plurality of approximate metadata sample values of the reconstructed metadata signal, wherein The metadata decoder 110 is configured to generate each of a plurality of approximate metadata sample values according to at least two of the plurality of first metadata sample values of the reconstructed metadata signal, for example, by using the approximate metadata sample value. Linear interpolation is produced, as shown in Figure 7. Painted.

According to the embodiment illustrated in FIG. 9, the metadata decoder 110 is configured to receive a plurality of differences for one of the at least one compressed metadata signal. Further, the metadata decoder 110 is configured to add each of the difference values to one of a plurality of approximate metadata sample values of the reconstructed metadata signal to obtain a plurality of second metadata samples of the reconstructed metadata signal. The value, and the reconstructed metadata signal is associated with the compressed metadata signal.

Since a difference has been received, all approximate metadata sample values are added to the approximate metadata sample values to obtain a plurality of second metadata sample values.

According to an embodiment, the approximate metadata sample value for which no difference is received is used as the second metadata sample value of the reconstructed metadata signal.

However, according to various embodiments, if no difference is received for the approximate metadata sample value, an approximate difference value is generated for the approximate difference sample value in accordance with the at least one received difference value, and the approximate metadata sample value and the approximate element are approximated The data sample values are added as described below.

According to the embodiment illustrated in FIG. 9, the received difference is summed with a plurality of metadata samples corresponding to the upsampled metadata signal. Thereby, when the difference has been transmitted, the difference of the corresponding interpolated metadata sample values can be correct, and if necessary, the correct metadata sample value can be obtained.

Referring to the metadata encoding of Figure 8, in the preferred embodiment, the number of bits used to encode the plurality of differences is less than the number of bits used to encode the metadata samples. These embodiments are based on the discovery that only a small change in subsequent (e.g., N) metadata samples over most of the time. For example, if a metadata sample value (eg, in 8-bits) is encoded, the metadata sample can take a difference from 256 different differences. Thus, subsequent (eg, N) metadata values typically vary slightly, and only the difference encoding (eg, in 5 bits) is considered sufficient. Therefore, even if only the difference is transmitted, the number of transmitted bits can be reduced.

In a preferred embodiment, at least one difference is transmitted, and each difference is encoded with a number of bits less than each of the metadata samples, wherein each difference is an integer.

According to an embodiment, the metadata encoder 110 is configured to encode the at least one metadata sample value of one of the at least one compressed metadata signal with a first number of bits, wherein at least one compressed metadata signal is used. One of each metadata sample value Represents an integer. In addition, the metadata encoder 110 is configured to encode at least one second difference of the plurality of second metadata samples and the second number of bits, wherein at least one difference of the plurality of second metadata samples Represents an integer in which the second number of bits is less than the first number of bits.

In an embodiment, the plurality of metadata samples may be encoded, for example, by azimuth in 8-bits, the azimuth being an integer and may be defined, for example, by the following equation: -90 Azimuth 90. Thus, the azimuth can be 181 different values, if it can be assumed that it is not much different from the subsequent (eg N) azimuth samples, eg no more than ±15, then 5 bits (25=32) may be sufficient to encode Multiple differences. If a plurality of different differences can represent a plurality of integer values, then it is determined that the plurality of different differences automatically convey additional values and are to be transmitted to an appropriate range of values.

For example, in the case where the first orientation value of the first source object is 60°, the subsequent plurality of azimuth values may vary between 45° and 75°. Furthermore, assuming that the second orientation value of the second source object is -30°, the subsequent plurality of orientation values will vary between -45° and -15°. For the difference between the two subsequent values of the second source object and the first source object, the plurality of different differences between the second orientation value and the first orientation value are between -15° and +15 Within a range of values, such that 5 bits are sufficient to encode each of a plurality of different difference values and such that a sequence of bits encoding a plurality of different differences has a plurality of second orientation values and a plurality of first orientation values Different difference.

In an embodiment, because no metadata samples are present in the compressed metadata signal, each difference is transmitted to the decoding side. Moreover, according to an embodiment, since no metadata samples are present in the compressed metadata signal, each difference is received and processed by the metadata decoder. However, some of the preferred embodiments illustrated in Figures 10 and 11 should be understood as different concepts.

Figure 10 is a diagram showing metadata encoding in accordance with another embodiment. According to various embodiments, the metadata encoder 210 can be used to implement the metadata encoding illustrated in FIG.

In some embodiments, as depicted in FIG. 10, the plurality of differences are used, for example, to determine each metadata sample value of the original metadata signal that is not included in the compressed metadata signal. For example, when the metadata sample values at the time points n=0 and n=N are included in the compressed metadata signal, but do not include the metadata sample values between the time points n=1 to n=N-1, Need to decide The difference between the time points n=1 and n=N-1.

However, according to the embodiment of FIG. 10, polygon approximation is then performed at 640. The metadata encoder 210 is used to determine which of the plurality of differences will be transmitted and to decide whether to transmit all of the differences.

For example, metadata encoder 210 may only transmit a plurality of differences having a difference greater than a threshold.

In another embodiment, when the ratio of the difference between the plurality of differences to the metadata sample value is greater than a threshold, the metadata encoder 210 may only transmit the difference.

In an embodiment, the metadata encoder 210 checks if the maximum absolute difference is greater than a threshold, and if so, transmits the absolute difference, otherwise, does not transmit any difference and ends the check. Continue to check the second largest difference and the third largest difference, etc. until all the differences are less than the threshold.

According to the embodiment, since not all the differences are necessarily transmitted, the metadata encoder 210 not only encodes it (in the plural numbers y1[k]...yN-1[k] in FIG. 10) a) a difference (size) and transmitting the original metadata signal associated with the difference (one of the plurality of values x1[k]...xN-1[k] in FIG. 10) Information on metadata sample values. For example, the metadata encoder can encode the point in time associated with the difference. For example, the metadata encoder may encode a value between 1 and N-1 to indicate a metadata sample value between 0 and N and transmit it in a compressed metadata signal associated with the difference. . Judging from the complex differences, the complex values x1[k]...xN-1[k]y1[k]...yN-1[k] listed on the output of the polygon approximation do not mean All values must be transmitted, and conversely, it means that no, one, some or all of the pairs of values will be transmitted.

In an embodiment, metadata encoder 210 may process portions (eg, N) of consecutive differences and approximate each portion by a polygonal process formed by the quantized variables of polygon points [xi, yi].

It is expected that the average of the variables of the polygonal points that must accurately approximate the difference signal is significantly smaller than N. In addition, [xi, yi] is a small integer value that will be encoded with the lower bits.

Figure 11 is a diagram showing metadata decoding in accordance with another embodiment. According to various embodiments, the metadata decoder 110 can be used to implement the metadata decoding illustrated in FIG.

In an embodiment, metadata decoder 110 receives some differences and adds the differences to the corresponding linearly interpolated metadata samples in 730.

In some embodiments, metadata decoder 110 simply adds the received difference value to the corresponding linearly interpolated metadata sample value within 730 and retains other linearities that do not receive any difference values. The inserted metadata sample value.

However, various embodiments implement another concept, as described below.

In accordance with such an embodiment, metadata decoder 110 receives a plurality of differences for one of the at least one compressed metadata signals, which means that each difference can be referred to as a "received difference." The received difference is designated as one of the approximate metadata samples of the reconstructed metadata signal, wherein the received difference is associated with the compressed metadata signal.

Referring to FIG. 9 already described, the metadata decoder 110 is configured to add each of the received plurality of differences to an approximate metadata sample value, the approximate metadata sample value and the received difference value. Associated. One of the plurality of second metadata samples of the reconstructed metadata signal is obtained by adding the received difference to its approximate metadata sample value.

However, for some approximate metadata sample values, typically no difference is received.

In some embodiments, when none of the plurality of received differences is associated with an approximate metadata sample value, the metadata decoder 110 may receive each of the approximate metadata samples for the reconstructed metadata signal, for example, based on the plurality of received metadata samples. The approximate difference is determined by at least one of the difference values, and the reconstructed metadata signal is associated with the compressed metadata signal.

In other words, for all approximate metadata samples, when no difference is received, the approximate difference is still generated based on at least one of the received differences.

The metadata decoder 110 is configured to add each of the plurality of approximate differences to the approximate metadata sample value of the approximate difference to obtain the other of the plurality of second metadata samples of the reconstructed metadata signal.

However, in another embodiment, for a plurality of metadata sample values, metadata decoder 110 performs linear interpolation in accordance with the plurality of differences received in step 740 for a plurality of differences that are not received. Make an approximation.

For example, if a first difference and a second difference are received, then a plurality of differences between the received plurality of differences can be approximated, such as by linear interpolation.

For example, when the first difference at time point n=15 has a difference d[15]=5 and when the second difference at time point n=18 has a difference d[18]=2, The complex differences of n=16 and d=17 can be linearly approximated as d[16]= and d[17]=3.

In another embodiment, when a plurality of metadata samples are included in the compressed metadata signal, the plurality of differences of the plurality of metadata samples are assumed to be zero, and the metadata decoder can be assumed to be 0. A plurality of metadata sample values are used to perform linear interpolation of a plurality of difference values that are not received.

For example, when a single difference d=8 at n=16 is transmitted and when the metadata samples at n=0 and n=32 are transmitted within the compressed metadata signal, then n=0 and The difference that is not transmitted on n=32 is assumed to be zero.

We assume that n represents time and that d[n] is the difference at time point n. Then: d[16]=8 (the difference received)

d[0]=0 (assumed difference as the metadata sample value present in z(k))

d[32]=0 (assumed difference, as the metadata sample value present in z(k))

Then a plurality of approximate metadata samples: d[1]=0.5; d[2]=1; d[3]=1.5; d[4]=2; d[5]=2.5;d[6]=3 ;d[7]=3.5;d[8]=4;d[9]=4.5;d[10]=5;d[11]=5.5;d[12]=6;d[13]=6.5; d[14]=7;d[15]=7.5;d[17]=7.5;d[18]=7;d[19]=6.5;d[20]=6;d[21]=5.5;d [22]=5;d[23]=4.5;d[24]=4;d[25]=3.5;d[26]=3;d[27]=2.5;d[28]=2;d[ 29]=1.5; d[30]=1; d[31]=0.5.

In an embodiment, the plurality of approximate differences received (in 730) are added to the corresponding linearly interpolated sample values.

The preferred embodiment is described below.

The (object) metadata encoder can encode a series of regular (sub)sampled track values using a look-ahead buffer of a given size N. Once the buffer is filled, the entire data block is encoded as well Transfer. The encoded object data can be composed of two parts, namely internally encoded object data and selective differential data containing the fine structure of each part.

The internally encoded object data contains quantized values z(k) that are sampled on a regular grid (length 1024 per 32 audio frames). A plurality of Boolean variables can be used to indicate for each object that a plurality of values are individually specified or used to indicate values applicable to all objects.

The decoder can extract coarse trajectories from the internally encoded object data by linear interpolation. The fine structure of the trajectory is given by differential data that contains the encoded difference between the input trajectory and the linear interpolation. For azimuth, elevation, and radius, polygon representation is combined with different quantization steps, resulting in a reduction in expected non-correlation.

Polygon representation can be obtained from a variation of the Douglas-Puck algorithm [10,11] that does not use recursion, where the Douglas-Puke algorithm differs from the original by using an additional interrupt loop. approximate.

The resulting plurality of polygon points can be encoded in the differential data portion using a variable character length that is specified within the data stream. Additional Boolean variants indicate a common encoding of the same values.

According to various embodiments, a plurality of object data frames and symbolic representations are described below.

In order to improve efficiency, a series of regular (sub)sampled track values are coded. The encoder can use a look-ahead buffer of a given size, and once the buffer is filled, the entire data block is encoded and transmitted. The encoded object data (eg, payload item metadata) may, for example, comprise two portions, internally encoded object data (first portion) and selective differential data (second portion).

For example, some or all of the following syntax can be used:

Internally encoded object data is described below in accordance with an embodiment: In order to support random reading of encoded object metadata, a complete and independent standard of all object metadata needs to be transmitted regularly. Here, it can be understood that the internal coded object data ("I frame") contains a plurality of quantized sample values on a regular grid (for example, the length of each 32 frames is 1024), wherein the I frame has The following syntax: after the current I frame, position_azimuth, position_elevation, position_radius, and gain_factor, these grammars specify a plurality of quantized sample values within the iframe_period frame.

Differential object data is described below in accordance with an embodiment.

A more accurate approximation is achieved by transmitting a smaller number of multiple sample point polygon approximations. Therefore, a very sparse three-dimensional matrix is transmitted, wherein the first dimension can be an object index, the second dimension can be a component of a plurality of metadata elements (azimuth, elevation, radius, gain) and the third dimension can be a complex number The frame index of the polygon sampling points. Without further measurement, the complex elements of the indicated matrix already require num_objects * num_components *(iframe_period-1) number of bits. The first step is to reduce the number of bits so that they can be added to four flags, which are used to indicate whether at least one value belongs to one of the four elements. For example, it can be expected that differential radius values or gain values will occur in only a few cases. The third dimension of the reduced three-dimensional matrix contains a vector having a plurality of iframe_period-1 elements. If only a small number of polygon points are expected, it is more efficient to parameterize the vector by a set of signal indices and the cardinality of the group. For example, for iframe_period of Nperiod=32 frames and a maximum of 16 polygon points, using this method would be more advantageous for Npoints<(32-log2(16))/log2(32)=5.6 polygon points. According to various embodiments, the following syntax for such an encoding scheme is employed:

The macro offset_data() encodes a plurality of locations (frame offsets) of a plurality of polygon points as a simple bit field or one of the concepts used above. A plurality of num_bits values allow for larger position skip coding, while the remainder of the difference data is edge coded with fewer characters.

In particular, in an embodiment, the plurality of macros may have, for example, the following meaning: According to an embodiment, the parameters of object_metadata() are defined as follows: has_differential_metadata indicates whether a differential object is present.

According to an embodiment, the parameters of intracoded_object_metadata() are defined as follows: ifperiod defines the number of frames between a plurality of independent frames.

Common_azimuth indicates whether the common azimuth is used for all objects.

Default_azimuth defines the value of the common azimuth.

Position_azimuth If there is no common azimuth value, then a value for each object is transmitted.

Common_elevation indicates whether the common elevation is used for all objects.

Default_elevation defines the value of the common elevation angle.

Position_elevation If there is no common elevation value, the value of each object is transmitted.

Common_radius indicates whether the common radius value is used for all objects.

Default_radius defines the value of the common radius.

Position_radius Transfers the value of each object if there is no common radius value.

Common_gain indicates whether the common gain value is used for all objects.

Default_gain defines the common gain factor value.

Gain_factor Transmits the value of each object if there is no common gain factor value.

Position_azimuth If there is only one object, this is its azimuth.

Position_elevation If there is only one object, this is its elevation angle.

Position_radius If there is only one object, this is its radius.

Gain_factor If there is only one object, this is its gain factor.

According to an embodiment, the parameters of differential_object_metadata() are defined as follows: bits_per_point is used to represent the number of bits needed for a polygon point.

Fixed_azimuth is used to indicate whether the azimuth value of all objects is a fixed flag.

Flag_azimuth A flag for each object that indicates whether the azimuth value has changed.

Nbits_azimuth is used to indicate how many bits are needed for the difference.

Differential_azimuth is the difference between the linear interpolation value and the actual value.

Fixed_elevation is used to indicate whether the elevation value of all objects is a fixed flag.

Flag_elevation is used to indicate each object flag for whether the elevation value has changed.

Nbits_elevation is used to indicate how many bits are needed for the difference.

Differential_elevation The difference between the linear interpolation and the actual value.

Fixed_radius is used to indicate whether the radius of all objects is a fixed flag.

Flag_radius is used to indicate each object flag for a change in radius.

Nbits_radius is used to indicate how many bits are needed for the difference.

Differential_radius The difference between the linear interpolation and the actual value.

Fixed_gain is used to indicate whether the gain factor of all objects is a fixed flag.

Flag_gain is used to indicate each object flag for whether the gain factor has changed.

Nbits_gain is used to indicate how many bits are needed for the difference.

Differential_gain The difference between the linear interpolation and the actual value.

According to an embodiment, the parameters of offset_data() are defined as follows: bitfield_syntax is used to indicate whether a vector having a plurality of polygon indices exists in a flag within the data stream.

Offset_bitfield The Boolean array contains a flag that is targeted at

Whether each point of iframe_period is a polygon point.

The number of npoints polygon points minus 1 (num_points=npoints+1).

Foffset The time division index of a polygon point within frame_period(frame_offset=foffset+1).

According to an embodiment, the metadata may be converted, for example, as a given location (eg, the indicated azimuth, elevation, and radius) of each source object on the defined timestamp.

In the prior art, on the one hand there is no variable technique for combining channels, and on the other hand object encoding enables the quality of the receivable source to be obtained at a low bit rate.

The 3D tone source side decoding system overcomes this limitation and is described below.

Figure 12 is a diagram showing a 3D sound source encoder in accordance with an embodiment of the present invention. The 3D sound source encoder is used to encode the sound source input data 101 to obtain the sound source output data 501. The 3D sound source encoder includes an input interface for receiving a plurality of sound source channels indicated by the CH and the plural indicated by the OBJ. One source object. In addition, the input interface 1100 illustrated in FIG. 12 additionally receives metadata associated with at least one of the plurality of source objects OBJ. In addition, the 3D sound source encoder includes a mixer 200 for mixing a plurality of objects and a plurality of channels to obtain a plurality of premixed channels, wherein each premixed channel includes one channel. Source data and sound source data of at least one object.

In addition, the 3D sound source encoder includes a core encoder 300 for core encoding its input data, and a metadata compressor 400 for compressing at least one of the plurality of sound source objects. A related metadata.

In addition, the 3D sound source encoder can include a mode controller 600 that controls the mixer, the core encoder, and/or an output interface 500 in one of a plurality of operating modes, wherein the core encoder is in the first mode system A plurality of source objects for encoding a plurality of source channels and received by the input interface 1100 without being affected by the mixer (ie, not being mixed by the mixer 200). However, in the second mode the mixer 200 is active and the core encoder encodes a plurality of mixed channels, i.e., the output produced by block 200. In the latter case, preferably, no more object data is encoded. Conversely, metadata indicating the location of a plurality of source objects has been used in the mixer 200 to translate the plurality of objects to the plurality of channels indicated by the metadata. on. In other words, the mixer 200 uses metadata associated with a plurality of source objects to pre-translate a plurality of source objects, and then the pre-translated plurality of source objects are mixed with the channels to obtain a mixed channel on the mixer output. . In this embodiment, any object may not have to be transmitted, and the source object may be applied to the compressed metadata and as an output of block 400. However, if not all of the objects of the input interface 1100 are mixed and only a certain number of objects are mixed, then only the objects that are not mixed and the associated metadata are still transferred to the encoder 300 or the metadata compressor, respectively. 400.

According to one of the various embodiments described above, the metadata compressor 400 in FIG. 12 is a metadata encoder 210 of the apparatus 250 for generating encoded sound source information. Moreover, in accordance with one of the various embodiments described above, the mixer 200 and the core encoder in FIG. 12 together comprise a source encoder 220 of the apparatus 250 for generating encoded source information.

Figure 14 is a diagram showing another embodiment of a 3D sound source encoder. The 3D sound source encoder in the figure further includes a SAOC encoder 800 for generating at least one transport channel and parameterized data from the spatial source object encoder input data. As shown in FIG. 14, the input data of the spatial sound source object encoder is an object that has not been processed by the pre-translator/mixer. Additionally, when the independent channel/object code is active in the first mode, the pre-translator/mixer is bypassed and all objects are input to the input interface 1100 encoded by the SAOC encoder 800.

Further, as shown in Fig. 14, preferably, the core encoder 300 is implemented as a USAC encoder, that is, as an encoder defined and standardized in the MPEG-USAC standard (USAC = Joint Speech and Source Coding). For the independent data type, all outputs of the 3D sound source encoder depicted in Fig. 14 are an MPEG 4 data stream having a container-like structure. Further, the metadata is indicated as "OAM" data, and the metadata compressor 400 in FIG. 12 corresponds to the OAM encoder 400 to obtain compressed OAM data input into the USAC encoder 300, as depicted in FIG. The USAC encoder 300 further includes an output interface for obtaining an MP4 output data stream having encoded channel/object data and compressed OAM data.

In accordance with one of the various embodiments described above, the OAM encoder 400 in FIG. 14 is a metadata encoder 210 of the apparatus 250 for generating encoded sound source information. Further, according to one of the above various embodiments, the SAOC encoder 800 in FIG. 14 And the USAC encoder 300 together form the source encoder 220 of the device 250 for generating encoded source information.

Figure 16 is a diagram showing another embodiment of a 3D sound source encoder. With respect to Figure 14, the SAOC encoder can be used to perform another encoding using the SAOC encoding algorithm, and the plurality of channels provided on the pre-translator/mixer 200 are not active in this mode, or SAOC encoding The device is used for SAOC encoding to add a plurality of pre-translation channels to the object. Therefore, the SAOC encoder 800 in FIG. 16 can operate on three different types of input data, that is, a plurality of channels do not have any preprocessed objects, a plurality of channels, and a plurality of pre-translated objects, or plural Separate objects. Further, preferably, another OAM decoder 420 is provided in FIG. 16 to cause the SAOC encoder 800 to process the same data on the encoder side, that is, the data obtained by distortion compression, instead of the original OAM data.

In Figure 16, the 3D source encoder can operate in multiple independent modes.

The 3D sound source encoder in Fig. 16 may additionally operate in the third mode except when the first mode and the second mode are described in the context of Fig. 12, when the pre-translation/mixer 200 is not active The core encoder generates at least one transport channel from the plurality of independent objects in the third mode. Additionally or additionally, when the pre-translation/mixer 200 corresponding to the mixer 200 in FIG. 12 is not active, the SAOC encoder generates at least one additional or additional transport from the plurality of original signals in the third mode. Channel.

Finally, when the 3D sound source encoder is used in the fourth mode, the SAOC encoder 800 can encode a plurality of channels of the plurality of pre-translated objects generated by the pre-translation/mixer. Therefore, in the fourth mode, since a plurality of channels and a plurality of objects are completely transmitted into a plurality of independent SAOC transport channels, the lowest bit rate application will provide good quality, and with FIG. 3 and The side code information indicated in Fig. 5 is associated as "SAOC-SI", and in the fourth mode, no compressed metadata is transmitted.

In accordance with one of the various embodiments described above, the OAM encoder 400 in FIG. 16 is a metadata encoder 210 of the apparatus 250 for generating encoded sound source information. Further, according to one of the above various embodiments, the SAOC encoder 800 and the USAC encoder 300 in Fig. 16 together constitute the sound source encoder 220 of the device 250 for generating encoded sound source information.

According to another embodiment, an apparatus for encoding sound source input data 101 to obtain sound source output data 501 is provided, comprising: an input interface 1100 for receiving a plurality of sound source channels, a plurality of sound source objects, and related to a plurality of sound source objects Metadata of at least one of the sound source objects; a mixer 200, which mixes the plurality of objects and the plurality of channels to obtain a plurality of pre-mixed channels, each of the plurality of pre-mixed channels includes a channel of one channel Data and sound source data of at least one object; a device 250 for generating encoded sound source information including a metadata encoder and a sound source encoder.

The source encoder 220 of the device 250 is a core encoder 300 that core encodes core encoder input data.

A metadata encoder 210 for generating means 250 for encoding source data is a metadata compressor 400 for compressing metadata relating to at least one of a plurality of source objects.

Figure 13 is a diagram showing a 3D sound source decoder in accordance with an embodiment of the present invention. The 3D sound source decoder receives the encoded sound source data as an input, that is, the data 501 of Fig. 12.

The 3D sound source decoder includes a metadata decompressor 1400, a core decoder 1300, a processor 1200, a mode controller 1600, and a post processor 1700.

Specifically, the 3D sound source decoder is configured to decode the encoded sound source data, and the input interface is configured to receive the encoded sound source data including the plurality of encoded channels and the plurality of encoded objects, in a specific mode, the compressed metadata system and A plurality of objects are associated.

In addition, the core decoder 1300 is configured to decode a plurality of code channels and a plurality of coded objects. Additionally, a metadata decompressor is used to decompress the compressed metadata.

In addition, the object processor 1200 is configured to use the plurality of decoded objects generated by the decompressed metadata processing core decoder 1300 to obtain a predetermined number of output channels including the object data and the plurality of decoded channels. The output channel is indicated at 1205 and then input to post processor 1700. The post processor 1700 is configured to convert a number of output channels 1205 into a particular output format, which may be a binary output format or a speaker output format, such as output formats such as 5.1 and 7.1.

Preferably, the 3D sound source decoder includes a mode controller 1600 for analyzing the encoded data to detect a mode indication. Therefore, the mode controller 1600 The system is connected to the input interface 1100 in FIG. However, a mode controller is not necessary here. Conversely, the adjustable source decoder can be preset by any other kind of control data, such as user input or any other control. Preferably, the 3D sound source decoder in FIG. 13 is controlled by mode controller 1600 and is used to bypass any object processor and feed a plurality of decoded channels to post processor 1700. When the second mode is applied to the 3D sound source encoder, that is, when the 3D sound source encoder of Fig. 12 operates in the second mode, only the pre-translation channel is received. In addition, when the first mode is applied to the 3D sound source encoder, that is, when the 3D sound source encoder has performed independent channel/object encoding, the object processor 1200 is not bypassed, and the plurality of decoded channels and The plurality of decoded objects are fed to the object processor 1200 along with the decompressed metadata generated by the metadata decompressor 1400.

Preferably, the indication that the first mode or the second mode is applied is included in the decoded sound source data, and the mode controller 1600 analyzes the decoded data to detect a mode indication. The first mode is used when the mode indication indicates that the encoded sound source data includes a plurality of coded channels and the plurality of coded objects; and the mode indication indicates that the coded source data does not contain any source objects (ie, only includes the 3D in FIG. 12) The second mode is used when a plurality of pre-translated channels are obtained by the sound source decoder.

In Fig. 13, in accordance with one of the various embodiments described above, the metadata decompressor 1400 is a metadata decoder 110 of the apparatus 100 for generating at least one source channel. Further, according to one of the above various embodiments, the core decoder 1300, the object processor 1200, and the post-processor 1700 in FIG. 13 together constitute the sound source decoder 120 of the device 100 for generating a plurality of sound source channels. .

Fig. 15 is a view showing an embodiment of the 3D sound source encoder with reference to Fig. 13, and the embodiment of Fig. 15 corresponds to the 3D sound source encoder of Fig. 14. The 3D sound source encoder in Fig. 15 includes a SAOC decoder 1800 in addition to the embodiment of the 3D sound source encoder in Fig. 13. In addition, the object processor 1200 of FIG. 13 is implemented as a separate object translator 1210 and a mixer 1220. The function of the object translator 1210 can also be implemented by the SAOC decoder 1800 according to different modes.

Additionally, post processor 1700 can be implemented as a binary translator 1710 or a format converter 1720. Alternatively, direct output of data 1205 of Figure 13 can be implemented, as depicted at 1730. Therefore, in order to have variability, it is preferable to use a larger number (for example, 22.2) Or the channel of 32) performs processing within the decoder, if a smaller format is required, followed by post processing. However, it is clear from the outset that only a small format (e.g., 5.1 format) is needed, preferably, as depicted by the shortcut 1727 of Figure 13 or Figure 6, an application across the SAOC decoder and/or the USAC decoder can be applied. A special control is taken to avoid unnecessary liter mixing operations and subsequent downmix operations.

In a preferred embodiment of the present invention, the object processor 1200 includes a SAOC decoder 1800 for decoding and using at least one transport channel and associated parameterized data output by the core decoder. The metadata is decoded to obtain a plurality of translated source objects. To this end, the OAM output is connected to block 1800.

In addition, the object processor 1200 is configured to translate a plurality of decoded objects output by the core decoder, which are not encoded in a plurality of SAOC transport channels, but are independently encoded in a plurality of typical singles indicated by the object translator 1210. Channel component. In addition, the decoder includes an output interface corresponding to output 1730 for outputting one of the mixer outputs to a plurality of speakers.

In another embodiment, the object processor 1200 includes a spatial sound source object codec 1800 for decoding at least one transport channel and associated parameterized side information representing a plurality of encoded sound source signals or a plurality of codes a source channel, wherein the spatial source codec is used to transcode the associated parameterized information and decompressed metadata to the transcoded parametric side information to enable direct translation of the output format, such as in An example defined by an earlier version of SAOC. The post processor 1700 is configured to use a plurality of decoded transport channels and transcoded parametric side information to calculate a plurality of source channels of the output format. The processing performed by the post processor may be similar to MPEG Surround processing or may be any other processing, such as BCC processing or the like.

In another embodiment, the object processor 1200 includes a spatial source object codec 1800 for transporting channels and parameterizing side information using a plurality of decodings (by the core decoder) for direct mixing of the output formats. And translation.

In addition, it is important that the object processor 1200 of FIG. 13 further includes a mixer 1220, when there is a mixture of a plurality of pre-translated objects and a plurality of channels (ie, when the mixer 200 of FIG. 12 is active), The mixer 1220 directly receives the number output by the USAC decoder 1300 According to and as an input. In addition, the mixer 1220 receives data that has not been SAOC decoded from an object translator that performs object translation. In addition, the mixer receives the SAOC decoder output data, that is, a plurality of SAOC translated objects.

Mixer 1220 is coupled to output interface 1730, binary translator 1710, and format converter 1720. Binary translator 1710 is used to translate a plurality of output channels into two binary channels using a head related transfer function or a binaural spatial impulse response (BRIR). The format converter 1720 is for converting a plurality of output channels into an output format having a number of channels smaller than the plurality of output channels 1205 of the mixer, and the format converter 1720 needs to reproduce information on the layout, for example 5.1 speakers, etc.

In accordance with one of the various embodiments described above, the OAM decoder 1400 in FIG. 15 is a metadata decoder 110 of the apparatus 100 for generating at least one source data. Further, according to one of the above various embodiments, the object translator 1210, the USAC decoder 1300, and the mixer 1220 in FIG. 15 together constitute the sound source decoder 120 of the apparatus 100 for generating at least one sound source channel. .

The 3D sound source decoder in Fig. 17 is different from the 3D sound source decoder in Fig. 15, except that the SAOC decoder can generate not only a plurality of translation objects but also a plurality of translation channels, in which case A 3D sound source decoder has been used in Figure 16, and the connection 900 between the plurality of channels/pre-translated objects and the SAOC encoder 800 input interface is active.

In addition, a vector reference amplitude shifting (VBAP) stage 1810 is used to receive information on the reproduction layout from the SAOC decoder and output the translation matrix to the SAOC decoder so that the SAOC decoder can be at 1205 at the terminal (ie, The high channel format of 32 speakers) provides a plurality of translation channels without any additional manipulation by the mixer.

Preferably, the VBAP block receives the decompressed OAM data to derive a plurality of translation matrices. More generally, it is preferred to have a reproduction layout and geometric information that the plurality of input signals should be translated to the position of the reproduction layout. The geometric input data can be OAM data for a plurality of object or channel position information, wherein a plurality of channels have been transmitted using SAOC.

However, if only a particular output interface is required, the VBAP stage 1810 has provided the required translation matrix for, for example, 5.1 output. SAOC decoder 1800 is executed Direct translation from the SAOC transport channel, associated parameter data, and decompressed metadata is directly translated into the desired output format without the interaction of the mixer 1220. However, when a particular blend is used between multiple modes, that is, several channels are SAOC encoded but not all channels are SAOC encoded; or several objects are SAOC encoded but not all objects are SAOC encoded; Only a certain number of pre-channel translated objects are SAOC decoded and the remaining channels are not processed by SAOC, then the mixer will take data from the individual input portions, ie directly from core decoder 1300, object translator 1210, and SAOC decoder 1800. put it together.

In Fig. 17, the metadata decoder 110 of the apparatus 100 for generating at least one source channel of one of the above embodiments is an OAM decoder 1400. Moreover, in Fig. 17, the sound source decoder 120 of the apparatus 100 for generating at least one sound source channel of one of the above-described plurality of embodiments is composed of an object translator 1210, a USAC decoder 1300, and a mixer 1220.

The present invention provides an apparatus for decoding encoded source data. The device for decoding the encoded sound source data comprises: an input interface 1100, configured to receive encoded sound source data, the encoded sound source data comprising a plurality of encoded channels, or a plurality of encoded objects, or compression of a plurality of objects Metadata; and a device 100 comprising a metadata decoder 110 and a tone channel generator 120 for generating at least one source channel as described above.

The metadata decoder 110 of the apparatus 100 for generating at least one source channel is a metadata decompressor 400 that decompresses the compressed metadata.

The source channel generator 120 of the apparatus 100 for generating at least one source channel includes a core decoder 1300 for decoding a plurality of code channels and a plurality of coded objects.

Moreover, the sound source channel generator 120 further includes an object processor 1200 that processes the plurality of decoded objects using the decompressed metadata to obtain a plurality of output channels 1205 including the sound source data from the plurality of objects and the plurality of decoded channels. The number.

In addition, the sound source channel generator 120 further includes a post processor 1700 that converts the number of the plurality of output channels 1205 into an output format.

Although some aspects have been described in the content of the device, it is clear that these aspects It also represents a description of the corresponding method, and the block or device corresponds to a method step or a method step. Likewise, the aspects described in the context of the method steps also represent a description of the corresponding blocks or items or features of the corresponding device.

The decompressed signal of the present invention can be stored on a digital storage medium or can be transmitted to a transmission medium (e.g., a wireless transmission medium) or a wired transmission medium (e.g., the Internet).

Embodiments of the invention may be implemented in hardware or on software, depending on the particular implementation requirements. This implements usability, digital storage media, such as floppy disks, DVDs, CDs, ROMs, PROM-EPROMs, EEPROMs or FLASH memories that store electronically readable control signals, which can be used with a programmable computer system ( Or can cooperate) to perform the above method.

Some embodiments in accordance with the present invention comprise a non-transitory data carrier having an electronically readable control signal that is capable of cooperating with a programmable computer system to perform one of the above methods.

In general, an embodiment of the present invention can be implemented as a computer program product having a program code that operates to perform one of the above methods when executed on a computer. For example, the code can be stored on a machine readable carrier.

Other embodiments include a computer program for performing one of the above methods, stored on a machine readable carrier.

In other words, an embodiment of the inventive method is therefore a computer program having a code that can execute one of the above methods when the computer program is executed on a computer.

Thus, another embodiment of the method of the present invention is a data carrier (either a digital storage medium or a computer readable medium) containing a computer program recorded to perform one of the above methods.

Thus, another embodiment of the method of the present invention is a data stream or a series of signals representing a computer program for performing one of the above methods. For example, the data stream or the serial signal can be configured to be transmitted via a data communication connection, such as through the Internet.

Another embodiment includes a processing device such as a computer or a programmable logic device for either performing one of the methods described above.

Another embodiment includes a computer having a computer program for performing one of the above methods.

In some embodiments, a programmable logic device, such as a field effect programmable gate array, can be used to perform some or all of the functions of the above methods. In some embodiments, in order to perform one of the above methods, the field effect programmable gate array can be mated to a microprocessor. Generally, this method can be preferably performed by any hardware device.

The above embodiments are merely illustrative of the principles of the invention. It is to be understood that the specific embodiments of the present invention are not intended to be limited In the case of the following patent application, various changes are made and are within the scope of the invention.

references:

[1] Peters, N., Lossius, T. and Schacher J. C., "SpatDIF: Principles, Specification, and Examples", 9th Sound and Music Computing Conference, Copenhagen, Denmark, Jul. 2012.

[2] Wright, M., Freed, A., "Open Sound Control: A New Protocol for Communicating with Sound Synthesizers", International Computer Music Conference, Thessaloniki, Greece, 1997.

[3] Matthias Geier, Jens Ahrens, and Sascha Spors. (2010), "Object-based audio reproduction and the audio scene description format", Org. Sound, Vol. 15, No. 3, pp. 219-227, December 2010.

[4] W3C, "Synchronized Multimedia Integration Language (SMIL 3.0)", Dec. 2008.

[5] W3C, "Extensible Markup Language (XML) 1.0 (Fifth Edition)", Nov. 2008.

[6] MPEG, "ISO/IEC International Standard 14496-3 - Coding of audio-visual objects, Part 3 Audio", 2009.

[7] Schmidt, J.; Schroeder, E. F. (2004), "New and Advanced Features for Audio Presentation in the MPEG-4 Standard", 116th AES Convention, Berlin, Germany, May 2004

[8] Web3D, "International Standard ISO/IEC 14772-1:1997 - The Virtual Reality Modeling Language (VRML), Part 1: Functional specification and UTF-8 encoding", 1997.

[9] Sporer, T. (2012), "Codierung räumlicher Audiosignale mit leicht-gewichtigen Audio-Objekten", Proc. Annual Meeting of the German Audiological Society (DGA), Erlangen, Germany, Mar. 2012.

[10] Ramer, U. (1972), "An iterative procedure for the polygonal approximation of plane curves", Computer Graphics and Image Processing, 1(3), 244-256.

[11] Douglas, D.; Peucker, T. (1973), "Algorithms for the reduction of the number of points required to represent a digitized line or its caricature", The Canadian Cartographer 10(2), 112-122.

[12] Ville Pulkki, “Virtual Sound Source Positioning Using Vector Base Amplitude Panning”; J. Audio Eng. Soc., Volume 45, Issue 6, pp. 456-466, June 1997.

100‧‧‧ device

110‧‧‧ metadata decoder

120‧‧‧Source channel generator, sound source decoder

Claims (19)

A device (100) for generating at least one source channel, wherein the device comprises: a metadata decoder (110) for receiving at least one compressed metadata signal, wherein each of the at least one compressed metadata signal The method includes a plurality of first metadata sample values, wherein the first metadata sample value of each of the at least one compressed metadata signal indicates information associated with at least one source object signal, wherein the metadata decoder (110) The method is configured to generate at least one re-established metadata signal, such that each of the at least one re-established metadata signal includes the plurality of first metadata samples of the at least one compressed metadata signal and further comprises a plurality of a binary data sample, wherein the metadata decoder (110) generates each of the reconstructed metadata signals according to at least two of the plurality of first metadata samples of the reconstructed metadata signal a second metadata sample value; and a sound source channel generator (120) generating the signal according to the at least one sound source object signal and the at least one reconstructed metadata signal One less source channel. The device (100) of claim 1, wherein the metadata decoder (110) generates each of the at least one by upsampling one of the at least one compressed metadata signals. And reconstructing the metadata signal, wherein the metadata decoder (110) performs linear interpolation according to at least two of the plurality of first metadata samples of the reconstructed metadata signal to generate each of the reconstructed metadata signals Each of the second metadata samples. The apparatus (100) of claim 1, wherein each of the reconstructed metadata signals includes the first metadata sample value of the at least one compressed metadata signal, and the reconstructed metadata signal is associated with the Corresponding to the compressed metadata signal, wherein the metadata decoder (110) generates the plurality of approximate metadata samples for the reconstructed metadata signal to generate the plural of each of the at least one reconstructed metadata signals a second metadata sample value, wherein the metadata decoder (110) is configured to generate at least two of the plurality of first metadata samples according to the reconstructed metadata signal to generate each of the plurality of approximate metadata samples a value, wherein the metadata decoder (110) is configured to receive a plurality of differences for the compressed metadata signal of the at least one compressed metadata signal, and to compare each of the plurality of differences with the One of the plurality of approximate metadata samples of the built-in data signal is added to obtain the plurality of second metadata samples of the reconstructed metadata signal, and the reconstructed metadata signal is associated with the compressed element Data signals are associated. The device (100) of claim 3, wherein the metadata decoder (110) is configured to receive the plurality of differences for the compressed metadata signal of the at least one compressed metadata signal, wherein Each of the plurality of differences is a received difference value assigned to one of the plurality of approximate metadata sample values of the reconstructed metadata signal associated with the compressed metadata signal, wherein the plurality of differences The data decoder (110) adds each of the received differences to the approximate metadata sample value associated with the received difference to obtain the plurality of second metadata of the reconstructed metadata signal. One of the sampled values, wherein when none of the plurality of received differences is associated with the approximate metadata sample value, the metadata decoder (110) is responsive to the at least one of the plurality of received differences An approximate difference value for each approximate metadata sample value of the reconstructed metadata signal associated with the compressed metadata signal, wherein the metadata decoder (110) is to each approximate Approximate value is added to the metadata of the sampled value of the difference approximation, to obtain the re-Chien wherein another of the plurality of data signals of the second sample values in the metadata. The device (100) of claim 1, wherein at least one of the at least one reconstructed metadata signals includes location information of one of the at least one source object signals, or includes the at least one audio source a zooming performance on the position information of one of the object signals, and wherein the sound source channel generator (120) is in accordance with the sound source object signal of the at least one sound source object signal and according to the position information to generate the at least one sound source sound At least one of the Tao. The device (100) of claim 1, wherein at least one of the at least one reconstructed metadata signals includes a volume of one of the at least one source object signals, or includes the at least one source object a zooming performance of the volume of one of the signals, and wherein the source channel generator (120) is in accordance with the source object signal of the at least one source object signal and according to the volume to generate the at least one source channel at least one. The device (100) of claim 1, wherein the device (100) receives random access information, wherein the random access information refers to the compressed information for each of the compressed metadata signals. One of the metadata signals accesses the signal portion, wherein at least one other signal portion of the metadata signal is not indicated by the random access information, and wherein the metadata decoder (110) is in accordance with the compressed metadata signal The first metadata sample value of the access signal portion, but not any other first metadata sample value of any other signal portion of the compressed metadata signal, to generate the at least one reconstructed metadata signal One. A device (250) for generating encoded sound source information, the encoded sound source information comprising at least one encoded sound source signal and at least one compressed metadata signal, wherein the device (250) comprises: a metadata decoder (210) for receiving At least one original metadata signal, wherein each of the at least one original metadata signal includes a plurality of metadata sample values, wherein the metadata sample value of each of the at least one original metadata signal is indicative of at least one The information associated with the source object signal of the source object signal, wherein the metadata encoder (210) is configured to generate the at least one compressed metadata signal such that each of the at least one compressed metadata signal includes the original metadata a first group of at least two of the metadata samples of one of the signals, and the plurality of metadata samples such that the compressed metadata signal does not include one of the plurality of original metadata signals Any of the other at least two of the second group of any metadata sample values, and a sound source encoder (220) for encoding the at least one The source object signal obtains the at least one encoded sound source signal. The device (250) of claim 8, wherein the metadata encoder (210) is configured to generate the at least one compressed metadata signal such that each of the at least one compressed metadata signal includes the at least one a first group of at least two of the metadata samples of one of the original metadata signals, the compressed metadata signal being associated with the original metadata signal, wherein the at least one original metadata Each of the metadata sample values included in the signal and also included in the compressed metadata signal is associated with the original metadata signal and is one of a plurality of first metadata sample values. , wherein one of the at least one original metadata signal is included and is not the compressed number of elements Each of the metadata sample values included in the signal is associated with the original metadata signal and is one of a plurality of second metadata samples, wherein the metadata encoder (210) is in accordance with Performing linear interpolation on at least two of the plurality of first metadata samples of one of the at least one original metadata signal for each of the plurality of original metadata signals The plurality of second metadata sample values produces an approximate metadata sample value, and wherein the metadata encoder (210) is for each of the at least one original metadata signal The data sampled value produces a difference such that the difference represents a difference between the second metadata sample value and the approximate metadata sample value of the second metadata sample value. The device (250) of claim 9, wherein the metadata encoder (210) is for the plurality of second metadata sample values of one of the at least one original metadata signal At least one of the plurality of differences determines whether each of the plurality of differences is greater than a threshold value. The device (250) of claim 9, wherein the metadata encoder (210) is configured to compare the at least one metadata sample value of the at least one compressed metadata signal with the first quantity. The bit is encoded, wherein each of the at least one compressed metadata signal represents an integer, wherein the metadata encoder (210) is configured to use the plurality of second The at least one second difference of the metadata sample value is encoded with the second number of bits, wherein the at least one difference of the plurality of second metadata sample values represents an integer, and wherein the second quantity The bit system is smaller than the first number of bits. The device (250) of claim 8, wherein at least one of the at least one original metadata signal includes location information of one of the at least one source object signal, or includes the at least one a scaled representation of the location information of one of the source object signals, and wherein the metadata encoder (210) generates the at least one compressed metadata signal in accordance with at least one of the at least one original metadata signal At least one of them. The device (250) of claim 8, wherein at least one of the at least one original metadata signal includes a volume of one of the at least one source object signal, or includes the at least one sound source a scaled representation of the volume of one of the object signals, and wherein the metadata encoder (210) generates at least one of the at least one compressed metadata signal in accordance with at least one of the at least one original metadata signal . A system, comprising: a device (250) according to any one of claims 8 to 13 for generating encoded source information, the encoded sound source information comprising at least one encoded sound source signal and at least one And a device (100) according to any one of claims 1 to 7 for receiving the at least one encoded sound source signal and the at least one compressed metadata signal, and Generating at least one source channel in accordance with the at least one encoded source signal and in accordance with the at least one compressed metadata signal. A method for generating at least one source channel, wherein the method includes receiving at least one compressed metadata signal, wherein each of the at least one compressed metadata signal comprises a plurality of first metadata samples, each of each The first metadata sample value of the at least one compressed metadata signal is indicative of information associated with the at least one source object signal, and the at least one reconstructed metadata signal is generated such that each of the at least one reconstructed metadata signal includes the at least one The plurality of first metadata sample values of one of the compressed metadata signals, and further comprising a plurality of second metadata sample values, wherein the step of generating the at least one reconstructed metadata signal includes the reconstructing the metadata signal according to the reconstructed metadata signal And at least two of the plurality of first metadata sample values to generate each of the second metadata sample values of each of the reconstructed metadata signals, and according to the at least one source object signal and according to the at least one Rebuilding the metadata signal generates the at least one source channel. A method for generating encoded source information, the encoded sound source information comprising at least one encoded sound source signal and at least one compressed metadata signal, wherein the method comprises: Receiving at least one original metadata signal, wherein each of the at least one original metadata signal includes a plurality of metadata sample values, wherein the metadata sample value of each of the at least one original metadata signal is indicative and at least The information associated with the source object signal of the source object signal generates the at least one compressed metadata signal such that each of the compressed metadata signals includes the plurality of metadata samples of the original metadata signal a first group of at least two, and a second group of at least two of the plurality of metadata samples that cause the compressed metadata signal to not include one of the plurality of original metadata signals Any metadata sample value, and encoding the at least one source object signal to obtain the at least one encoded sound source signal. A computer program for implementing a computer program as set forth in claim 15 or 16 when the computer program is executed on a computer or a signal processor. An apparatus for encoding sound source input data (101) to obtain sound source output data (501), comprising: an input interface (1100) for receiving a plurality of sound source channels, a plurality of sound source objects, and relating to at least one Metadata of a plurality of sound source objects; a mixer (200) mixing the plurality of objects and the plurality of channels to obtain a plurality of pre-mixed channels, each of the plurality of pre-mixed channels comprising one channel The sound source data and the sound source data of the at least one object; and the device (250) according to any one of claims 8 to 13, wherein any one of the claims 8th to 13th is applied for. The sound source encoder (220) of the device (250) is a core encoder (300) for core coding the core encoder input data, and the application of the patent scopes 8 to 13 The metadata encoder (210) of the device (250) is a metadata compressor (400) that compresses metadata about the at least one of the plurality of source objects. An apparatus for decoding encoded source data, comprising: an input interface (1100), configured to receive the encoded sound source data, the encoded sound source data comprising a plurality of encoded channels, or a plurality of encoded objects, or related The plurality of objects And a device (100) according to any one of claims 1 to 7, wherein the device (100) according to any one of claims 1 to 7 The metadata decoder (110) is a metadata decompressor (400) for decompressing the compressed metadata; wherein the device of any one of claims 1 to 7 is 100) The sound source channel generator (120) includes a core decoder (1300) for decoding the plurality of code channels and the plurality of coded objects, wherein the sound source channel generator (120) further comprises An object processor (1200) that processes the plurality of decoded objects using decompressed metadata to obtain a plurality of output channels (1205) including sound source data from the object and the decoded channel. The number, and wherein the source channel generator (120) further includes a post processor (1700) that converts the number of the plurality of output channels (1205) into an output format.