Abstract: Higher Order Ambisonics represents three-dimensional sound independent of a specific loudspeaker set-up. However, transmission of an HOA representation results in a very high bit rate. Therefore compression with a fixed number of channels is used, in which directional and ambient signal components are processed differently. For coding, portions of the original HOA representation are predicted from the directional signal components. This prediction provides side information which is required for a corresponding decoding. By using some additional specific purpose bits, a known side information coding processing is improved in that the required number of bits for coding that side information is reduced on average.

Abstract: Audio content coded for a reference speaker configuration is downmixed to downmix audio content coded for a specific speaker configuration. One or more gain adjustments are performed on individual portions of the downmix audio content coded for the specific speaker configuration. Loudness measurements are then performed on the individual portions of the downmix audio content. An audio signal that comprises the audio content coded for the reference speaker configuration and downmix loudness metadata is generated. The downmix loudness metadata is created based at least in part on the loudness measurements on the individual portions of the downmix audio content.

Abstract: A device for processing audio signals includes an interchannel phase difference (IPD) mode selector and an IPD estimator. The IPD mode selector is configured to select an IPD mode based on at least a strength value associated with a temporal misalignment between a first audio signal and a second audio signal. The IPD estimator is configured to determine IPD values based on the first audio signal and the second audio signal. The IPD values have a resolution corresponding to the selected IPD mode.

Abstract: A decoder for generating a frequency enhanced audio signal, includes: a feature extractor for extracting a feature from a core signal; a side information extractor for extracting a selection side information associated with the core signal; a parameter generator for generating a parametric representation for estimating a spectral range of the frequency enhanced audio signal not defined by the core signal, wherein the parameter generator is configured to provide a number of parametric representation alternatives in response to the feature, and wherein the parameter generator is configured to select one of the parametric representation alternatives as the parametric representation in response to the selection side information; and a signal estimator for estimating the frequency enhanced audio signal using the parametric representation selected.

Abstract: The encoding and decoding of HOA signals using Singular Value Decomposition includes forming based on sound source direction values and an Ambisonics order corresponding ket vectors (|Y(?s)) of spherical harmonics and an encoder mode matrix (?OxS). From the audio input signal (|x(?s)) a singular threshold value (?s) determined. On the encoder mode matrix a Singular Value Decomposition is carried out in order to get related singular values which are compared with the threshold value, leading to a final encoder mode matrix rank (rfine). Based on direction values (?l) of loudspeakers and a decoder Ambisonics order (Nl), corresponding ket vectors (|Y(?l)) and a decoder mode matrix (?OxL) are formed. On the decoder mode matrix a Singular Value Decomposition is carried out, providing a final decoder mode matrix rank (rfind).

Abstract: The present subject matter provides a technical solution to the technical problems facing sound localization by separating sounds and reproducing the separated sounds using a set of loudspeakers and a set of headphones. A general soundtrack that is meant to be experienced throughout the room would play through the loudspeakers, and specific sounds that are meant to be experienced near the listener would be played through a binaural representation in the headphones. The headphones may be selected to avoid occluding the ear, allowing sound produced at the loudspeakers to be heard clearly. This separation and reproduction of sounds using a combination of a loudspeaker and headphone provides a technical solution to the technical problem facing typical surround sound systems by localizing sounds for listeners in any location within a room. This improves reproduction accuracy of location-specific audio objects, including audio objects above or below a coplanar speaker configuration.

Abstract: Higher Order Ambisonics represents three-dimensional sound independent of a specific loudspeaker set-up. However, transmission of an HOA representation results in a very high bit rate. Therefore compression with a fixed number of channels is used, in which directional and ambient signal components are processed differently. For coding, portions of the original HOA representation are predicted from the directional signal components. This prediction provides side information which is required for a corresponding decoding. By using some additional specific purpose bits, a known side information coding processing is improved in that the required number of bits for coding that side information is reduced on average.

Abstract: From an HOA signal representation (c(t)) of a sound field having an order of N and a number 0=(N+1)2 of coefficient sequences a mezzanine HOA signal representation (wMEZZ(t)) is generated that consists of an arbitrary number I<0 of virtual loudspeaker signals WMEZZ1(t), WMEZZ,2(t), . . . , WMEZZ,I(t). 0 directions are computed which are nearly uniformly distributed on the unit sphere. The mode vectors with respect to these directions are linearly weighted for constructing a matrix, of which the pseudo-inverse is used for multiplying the HOA signal representation (c(t)) in order to form (11) the mezzanine HOA signal representation (WME.ZZ(t)).

Abstract: A method for using a loudspeaker array that is housed in a loudspeaker cabinet to present audio content to a listener in a room includes receiving (1) an audio channel that includes audio content and (2) acoustical characteristics of the room. The method also produces (1) a first beamformer input signal from the audio channel and (2) a second beamformer input signal and a third beamformer input signal by decorrelating the audio channel and adjusting the audio channel in accordance with the acoustical characteristics of the room. The second and third beamformer input signals are different de-correlated versions of the audio channel. The method also generates driver signals from the first, second, and third beamformer input signals to drive the loudspeaker array to produce a main beam, a first ambient beam, and a second ambient beam, respectively. Other embodiments are also described and claimed.

Abstract: Higher Order Ambisonics represents three-dimensional sound independent of a specific loudspeaker set-up. However, transmission of an HOA representation results in a very high bit rate. Therefore compression with a fixed number of channels is used, in which directional and ambient signal components are processed differently. For coding, portions of the original HOA representation are predicted from the directional signal components. This prediction provides side information which is required for a corresponding decoding. By using some additional specific purpose bits, a known side information coding processing is improved in that the required number of bits for coding that side information is reduced on average.

Abstract: A multi-channel decoder for generating a binaural signal from a downmix signal using upmix rule information on an energy-error introducing upmix rule for calculating a gain factor based on the upmix rule information and characteristics of head related transfer function based filters corresponding to upmix channels. The one or more gain factors are used by a filter processor for filtering the downmix signal so that an energy corrected binaural signal having a left binaural channel and a right binaural channel is obtained.

Abstract: A multi-channel decoder for generating a binaural signal from a downmix signal using upmix rule information on an energy-error introducing upmix rule for calculating a gain factor based on the upmix rule information and characteristics of head related transfer function based filters corresponding to upmix channels. The one or more gain factors are used by a filter processor for filtering the downmix signal so that an energy corrected binaural signal having a left binaural channel and a right binaural channel is obtained.

Abstract: There are two representations for Higher Order Ambisonics denoted HOA: spatial domain and coefficient domain. The invention generates from a coefficient domain representation a mixed spatial/coefficient domain representation, wherein the number of said HOA signals can be variable. A vector of coefficient domain signals is separated into a vector of coefficient domain signals having a constant number of HOA coefficients and a vector of coefficient domain signals having a variable number of HOA coefficients. The constant-number HOA coefficients vector is transformed to a corresponding spatial domain signal vector. In order to facilitate high-quality coding, without creating signal discontinuities the variable-number HOA coefficients vector of coefficient domain signals is adaptively normalized and multiplexed with the vector of spatial domain signals.

Abstract: Audio content coded for a reference speaker configuration is downmixed to downmix audio content coded for a specific speaker configuration. One or more gain adjustments are performed on individual portions of the downmix audio content coded for the specific speaker configuration. Loudness measurements are then performed on the individual portions of the downmix audio content. An audio signal that comprises the audio content coded for the reference speaker configuration and downmix loudness metadata is generated. The downmix loudness metadata is created based at least in part on the loudness measurements on the individual portions of the downmix audio content.

Abstract: Disclosed is an audio signal processing device for processing an audio signal. The audio signal processing device includes a receiving unit configured to receive the audio signal; a processor configured to determine whether to render the audio signal by reflecting a location of a sound image simulated by the audio signal on the basis of metadata for the audio signal, and render the audio signal according to a result of the determination; and an output unit configured to output the rendered audio signal.

Abstract: The present disclosure regards a binaural hearing system configured to receive sound signals from the environment having two hearing instruments to be worn on respective sides of the head of a user and to generate a binaural signal using the received sound signals of both hearing instruments.

Abstract: According to an aspect of an embodiment, an audio generation method includes: receiving an audio signal through at least one microphone; generating an input channel signal corresponding to each of the at least one microphone based on the received audio signal; generating a virtual input channel signal based on the input channel signal; generating additional information including reproduction locations of the input channel signal and the virtual input channel signal; and transmitting a multi-channel audio signal and the additional information, the multi-channel audio signal including the input channel signal and the virtual input channel signal.

Abstract: The encoding and decoding of HOA signals using Singular Value Decomposition includes forming based on sound source direction values and an Ambisonics order corresponding ket vectors (|Y(?s)) of spherical harmonics and an encoder mode matrix (?O×S). From the audio input signal (|x(?s)) a singular threshold value (?S) determined. On the encoder mode matrix a Singular Value Decomposition is carried out in order to get related singular values which are compared with the threshold value, leading to a final encoder mode matrix rank (rfins). Based on direction values (?l) of loudspeakers and a decoder Ambisonics order (Nl), corresponding ket vectors (|Y(?l)) and a decoder mode matrix (?O×L) are formed. On the decoder mode matrix a Singular Value Decomposition is carried out, providing a final decoder mode matrix rank (rfind).

Abstract: Higher Order Ambisonics represents three-dimensional sound independent of a specific loudspeaker set-up. However, transmission of an HOA representation results in a very high bit rate. Therefore compression with a fixed number of channels is used, in which directional and ambient signal components are processed differently. For coding, portions of the original HOA representation are predicted from the directional signal components. This prediction provides side information which is required for a corresponding decoding. By using some additional specific purpose bits, a known side information coding processing is improved in that the required number of bits for coding that side information is reduced on average.

Abstract: A signaling processor is provided. The signaling processor includes a frequency domain processing module configured to generate a cut-off frequency of an input signal and to generate level information for adjusting a level of a high frequency recovery signal and a time domain processing module configured to receive the cut-off frequency and the level information from the frequency domain processing module, to generate a signal having a frequency greater than or equal to the cut-off frequency using part of a signal of a frequency lower than the cut-off frequency in the input signal, to generate the high frequency recovery signal by adjusting a level of the generated signal using the level information, and to synthesize the high frequency recovery signal with the input signal.

Abstract: Audio content coded for a reference speaker configuration is downmixed to downmix audio content coded for a specific speaker configuration. One or more gain adjustments are performed on individual portions of the downmix audio content coded for the specific speaker configuration. Loudness measurements are then performed on the individual portions of the downmix audio content. An audio signal that comprises the audio content coded for the reference speaker configuration and downmix loudness metadata is generated. The downmix loudness metadata is created based at least in part on the loudness measurements on the individual portions of the downmix audio content.

Abstract: Provided is a method and apparatus for encoding/decoding an audio signal. Sections which are not used to output noise components near important spectral components and sub-bands which are not used to output noise components, are determined to be encoded or decoded, so that the efficiency of encoding and decoding an audio signal increases, and sound quality can be improved using less bits.

Abstract: Diffuse or spatially large audio objects may be identified for special processing. A decorrelation process may be performed on audio signals corresponding to the large audio objects to produce decorrelated large audio object audio signals. These decorrelated large audio object audio signals may be associated with object locations, which may be stationary or time-varying locations. For example, the decorrelated large audio object audio signals may be rendered to virtual or actual speaker locations. The output of such a rendering process may be input to a scene simplification process. The decorrelation, associating and/or scene simplification processes may be performed prior to a process of encoding the audio data.

Abstract: The present document relates to the field of encoding and decoding of audio. In particular, the present document relates to encoding and decoding of an audio scene comprising audio objects. A method (400) for encoding metadata relating to a plurality of audio objects (106a) of an audio scene (102) is described. The metadata comprises a first set (114, 314) of metadata and a second set (104) of metadata. The first and second sets (104, 114, 314) of metadata comprise one or more data elements which are indicative of a property of an audio object (106a) from the plurality of audio objects (106a) and/or of a downmix signal (112) derived from the plurality of audio objects (106a). The method (400) comprises identifying (401) a redundant data element which is common to the first and second sets (104, 114, 314) of metadata.

Abstract: Disclosed are a multi-channel audio signal processing method and a multi-channel audio signal processing apparatus. The multi-channel audio signal processing method may generate N channel output signals from N/2 channel downmix signals based on an N-N/2-N structure.

Abstract: In general, techniques are directed to intermediate compression of higher order ambisonic audio data. For example, a device comprising a processor and a memory may be configured to perform the techniques. The memory may be configured to store an intermediately formatted audio data generated as a result of an intermediate compression of higher order ambisonic audio data. The one or more processors may be configured to process the intermediately formatted audio data.

Abstract: A method for selecting a subset of channels of (e.g., determined from) at least a segment of a multichannel audio program for watermarking and watermarking the selected subset of channels, and a system or device configured to implement any embodiment of the method, or including a buffer which stores at least one frame or other segment of a multichannel audio program generated by any embodiment of the method or steps thereof. Some embodiments generate watermarking metadata during program creation including by analyzing audio content to be included in segments of a multichannel program, determining at least one watermark suitability value for each channel of each of the segments, and including the watermark suitability values (or watermarking data determined therefrom) as metadata in the program. Some embodiments are implemented by a playback system which determines the selected subset of channels to be watermarked.

Abstract: Audio content coded for a reference speaker configuration is downmixed to downmix audio content coded for a specific speaker configuration. One or more gain adjustments are performed on individual portions of the downmix audio content coded for the specific speaker configuration. Loudness measurements are then performed on the individual portions of the downmix audio content. An audio signal that comprises the audio content coded for the reference speaker configuration and downmix loudness metadata is generated. The downmix loudness metadata is created based at least in part on the loudness measurements on the individual portions of the downmix audio content.

Abstract: In general, techniques are described for transforming spherical harmonic coefficients. A device comprising one or more processors may perform the techniques. The processors may be configured to parse the bitstream to determine transformation information describing how the sound field was transformed to reduce a number of the plurality of hierarchical elements that provide information relevant in describing the sound field. The processors may further be configured to, when reproducing the sound field based on those of the plurality of hierarchical elements that provide information relevant in describing the sound field, transform the sound field based on the transformation information to reverse the transformation performed to reduce the number of the plurality of hierarchical elements.

Abstract: Disclosed is a non-transitory computer readable storage medium which receives, by an audio decoder (operating in a specific playback environment different from a reference channel configuration), an audio signal for the reference channel configuration. The audio signal includes audio sample data and encoder-generated loudness metadata which includes a plurality of portions of loudness metadata for a plurality of playback environments. The plurality of portions of loudness metadata includes one or more respective portions of loudness metadata for each playback environment in the plurality of playback environments. The medium also selects one or more portions of specific loudness metadata (based on the specific playback environment), from among the plurality of portions of loudness metadata for the plurality of playback environments.

Abstract: Diffuse or spatially large audio objects may be identified for special processing. A decorrelation process may be performed on audio signals corresponding to the large audio objects to produce decorrelated large audio object audio signals. These decorrelated large audio object audio signals may be associated with object locations, which may be stationary or time-varying locations. For example, the decorrelated large audio object audio signals may be rendered to virtual or actual speaker locations. The output of such a rendering process may be input to a scene simplification process. The decorrelation, associating and/or scene simplification processes may be performed prior to a process of encoding the audio data.

Abstract: A device having first means for directing a sound wave to a first side of a target head location; a first means for encircling the first directing means, being configured to phase focus the sound wave to a proximate center point of a left ear of the listener; a second means for directing the sound wave to a second side of the target head location of the listener; a second means for encircling the second directing means, being configured to phase focus the sound wave to a proximate center point of a right ear of the listener; and means for encapsulating the first and second encircling means, said encapsulating means being configured to proximately surround the target head location of the listener.

Abstract: An encoder (501) generates data representing an audio scene by a first downmix and data characterizing audio objects. In addition, a direction dependent diffuseness parameter indicative of a degree of diffuseness of a residual downmix is provided where the residual downmix corresponds to a downmix of audio components of the audio scene with the audio objects being extracted. A rendering apparatus (503) comprises a receiver (701) receiving the data from the encoder (501). A circuit (703) generates signals for a spatial speaker configuration from the audio objects. A transformer (709) generates non-diffuse sound signals for the spatial speaker configuration by applying a first transformation to the residual downmix and another transformer (707) generates signals for the spatial speaker configuration by applying a second transformation to the residual downmix by applying a decorrelation to the residual downmix. The transformations are dependent on the direction dependent diffuseness parameter.

Abstract: Provided is a method and apparatus for encoding/decoding an audio signal. Sections which are not used to output noise components near important spectral components and sub-bands which are not used to output noise components, are determined to be encoded or decoded, so that the efficiency of encoding and decoding an audio signal increases, and sound quality can be improved using less bits.

Abstract: Audio content coded for a reference speaker configuration is downmixed to downmix audio content coded for a specific speaker configuration. One or more gain adjustments are performed on individual portions of the downmix audio content coded for the specific speaker configuration. Loudness measurements are then performed on the individual portions of the downmix audio content. An audio signal that comprises the audio content coded for the reference speaker configuration and downmix loudness metadata is generated. The downmix loudness metadata is created based at least in part on the loudness measurements on the individual portions of the downmix audio content.

Abstract: Received audio data may include a first set of frequency coefficients and a second set of frequency coefficients. Spatial parameters for at least part of the second set of frequency coefficients may be estimated, based at least in part on the first set of frequency coefficients. The estimated spatial parameters may be applied to the second set of frequency coefficients to generate a modified second set of frequency coefficients. The first set of frequency coefficients may correspond to a first frequency range (for example, an individual channel frequency range) and the second set of frequency coefficients may correspond to a second frequency range (for example, a coupled channel frequency range). Combined frequency coefficients of a composite coupling channel may be based on frequency coefficients of two or more channels. Cross-correlation coefficients, between frequency coefficients of a first channel and the combined frequency coefficients, may be computed.

Abstract: Provided are a multi-channel audio down-mixing method and apparatus for selecting down-mix target channels based on a calculation of correlations between channels and then down-mixing the down-mix target channels. The method includes: calculating correlations between channels of multi-channel audio; selecting a first channel and a second channel, among the channels of the multi-channel audio, that are to be down-mixed, based on the calculated correlations; and down-mixing the selected first channel and the selected second channel.

Abstract: A decoding system reconstructs an audio signal based on an input signal representing the audio signal by parametric coding or by n discretely coded channels. Parametric decoding proceeds on the basis of a core signal and mixing parameters controlling a spatial synthesis stage, which is supplied with a downmix signal. A controller is responsible for controlling the components of the decoding system, whether in steady-state parametric mode, steady-state discrete decoding mode and transitions between these. In defective frames of the input signal, which do not allow the mixing parameters to be decoded, the controller is configured to perform various error handling procedures including: parametric decoding using previous values of the mixing parameters; continuing parametric decoding for a limited duration, and/or outputting the core signal without spatial synthesis.

Abstract: The present invention proposes a method for operating a binaural hearing system with two hearing devices (1, 1?) operationally interconnected by means of a bidirectional link (8) which improves hearing perception in windy listening situations. The method comprises determining the level of wind noise present at each of the two hearing devices (1, 1?) and sending the audio signal picked-up at the first hearing device (1) to the second hearing device (1?) via the link (8) and then providing an output signal derived from the received signal to the electrical-to-mechanical output converter (3?) of the second hearing device (1?) if the level of wind noise at the second hearing device (1?) exceeds the level of wind noise at the first hearing device (1) by a pre-set threshold value. Furthermore, a binaural hearing system capable of performing such a method is given.

Abstract: The invention relates to a method for determining an encoding parameter for an audio channel signal of a multi-channel audio signal, the method comprising: determining for the audio channel signal a set of functions from the audio channel signal and a reference audio signal; determining a first set of encoding parameters based on a smoothing of the set of functions with respect to a frame sequence of the multi-channel audio signal, the smoothing being based on a first smoothing coefficient; determining a second set of encoding parameters based on a smoothing of the set of functions with respect to the frame sequence of the multi-channel audio signal, the smoothing being based on a second smoothing coefficient; and determining the encoding parameter based on a quality criterion with respect to the first set of encoding parameters and/or the second set of encoding parameters.

Abstract: The invention relates to a method for determining an encoding parameter for an audio channel signal of a multi-channel audio signal, the method comprising: determining a frequency transform of the audio channel signal; determining a frequency transform of a reference audio signal; determining inter channel differences for at least each frequency sub-band of a subset of frequency sub-bands, each inter channel difference indicating a phase difference or time difference between a band-limited signal portion of the audio channel signal and a band-limited signal portion of the reference audio signal in the respective frequency sub-band the inter-channel difference is associated to; determining a first average based on positive values of the inter-channel differences and determining a second average based on negative values of the inter-channel differences; and determining the encoding parameter based on the first average and on the second average.

Abstract: An apparatus for generating a decorrelated signal having a receiving unit for receiving phase information, a transient separator, a transient decorrelator, a second decorrelator and a combining unit, wherein the transient separator is adapted to separate an input signal into a first signal component and into a second signal component such that the first signal component has transient signal portions of the input signal and such that the second signal component has non-transient signal portions of the input signal. The transient decorrelator is adapted to apply the phase information received by the receiving unit to a transient signal component.

Abstract: The invention relates to the technical field of audio encoding and/or decoding technologies, and thus concerns an overall encoding procedure and associated decoding procedure. The encoding procedure involves at least two signal encoding processes (S1-S3) operating on signal representations of a set of audio input channels, as well as residual encoding (S7-S8). It also involves a dedicated process (S4-S6) to estimate and encode energies of the audio input channels. Each encoding process is associated with a corresponding decoding process. In the overall decoding procedure the decoded signals from each encoding process are preferably combined such that the output channels are close to the input channels in terms of energy and/or quality. Normally, the combination step also adapts to the possible loss of one or more signal representation in part or in whole, such that the energy and quality is optimized with the signals at hand in the decoder. In this way, the overall quality of the output channels is improved.

Abstract: Methods and systems for providing synthetic audio placement are described herein. In different aspects, the methods and systems may include a connection port to receive a plurality of remote audio signals included in a conference call, at least one of the plurality of remote audio connections being a single channel audio signal, a processor configured with the connection port to provide a sound field for placing the plurality of audio signals, the sound field providing synthetic audio placement of each of the plurality of remote audio signals, and a rendering output configured with the processor to project the plurality of remote audio within the sound field to a listener.

Abstract: An amplifier unit suitable for a wide variety of uses, which can flexibly support diverse sound recording scenarios, and a mobile electronic device provided with the amplifier unit. An amplifier unit has microphone/line input terminals respectively provided on the topside surface and the left side surface. When a connector is connected to a mobile electronic device, stereophonic sound is recorded using one of the terminals when the mobile electronic device is held in portrait orientation, and using the other terminal when the mobile electronic device is held in landscape orientation. It is possible to switch each of audio signals coming through the two terminals between monophonic and stereophonic.

Abstract: The application relates to HFR (High Frequency Reconstruction/Regeneration) of audio signals. In particular, the application relates to a method and system for performing HFR of audio signals having large variations in energy level across the low frequency range which is used to reconstruct the high frequencies of the audio signal. A system configured to generate a plurality of high frequency subband signals covering a high frequency interval from a plurality of low frequency subband signals is described.

Abstract: A coding apparatus which suppresses an extreme increase in a bit rate, includes: a downmixing and coding unit (301) that downmixes audio signals that have been provided, to reduce the number of channels to be fewer than the number of the provided audio signals, and to code the downmix signals; an object parameter extracting unit (304) that extracts parameters indicating correlation between the audio signals; and a multiplexing circuit (309) that multiplexes the extracted parameters with the generated downmix coded signals. The object parameter extracting unit (304) includes: an object classifying unit (305) that classifies each of the provided audio signals into a predetermined one of types based on audio characteristics; and an object parameter extracting circuit (308) that extracts parameters using a temporal granularity and a frequency granularity each of which is determined for a corresponding one of the types.

Abstract: The invention relates to an audio signal synthesizer, the audio signal synthesizer comprises a transformer for transforming the down-mix audio signal into frequency domain to obtain a transformed audio signal; a signal generator for generating a first auxiliary signal, for generating a second auxiliary signal, and for generating a third auxiliary signal upon the basis of the transformed audio signal; a de-correlator for generating a first de-correlated signal, and for generating a second de-correlated signal from the third auxiliary signal, the first de-correlated signal and the second de-correlated signal being at least partly de-correlated; and a combiner for combining the first auxiliary signal with the first de-correlated signal to obtain a first audio signal, and for combining the second auxiliary signal with the second de-correlated signal to obtain the second audio signal, the first audio signal and the second audio signal forming the multi-channel audio signal.

Abstract: An apparatus and method for encoding and decoding a spatial parameter are provided. A spatial parameter encoding apparatus may encode a spatial parameter using a correlation between spatial parameters indicating a characteristic relationship between channels of a multi-channel audio signal, so that the multi-channel audio signal may be efficiently encoded.

Abstract: An encoder for a multi-channel audio signal which comprises a down-mixer (201, 203, 205) for generating a down-mix as a combination of at least a first and second channel signal weighted by respectively a first and second weight with different amplitudes for at least some time-frequency intervals. Furthermore, a circuit (201, 203, 209) generates up-mix parametric data characterizing a relationship between the channel signals as well as characterizing the weights. A circuit generates weight estimates for the encoder weights from the up-mix parametric data; and comprises an up-mixer (407) which recreates the multi channel audio signal by up-mixing the down-mix in response to the up-mix parametric data, the first weight estimate and the second weight estimate. The up-mixing is dependent on the amplitude of at least one of the weight estimate(s).