Coherent calculation for stereo discontinuous transmission (DTX)

Abstract

Enabling generation of comfortable noise in an encoder using estimated coherence parameters in a network using discontinuous transmission (DTX) comprises: receiving a time domain audio input comprising an audio input signal; and processing the input signals frame by frame by encoding active content of each input signal at a first bit rate until an inactive period is detected in the input signal; switching encoding from active encoding to inactive encoding during the inactive period in order to encode background noise at a second bit rate; estimating a coherence parameter during an inactive period based on low-pass filtering of the cross-spectrum or an average of the cross-spectrum, including re-initializing a low-pass filter state of the cross-spectrum based on a coherence parameter from a previous inactive period; encoding the estimated coherence parameters; and initiating transmission of the encoded active content, background noise, and coherence parameters to a decoder.

Description

Technical Field

[0001] The present disclosure relates generally to communications, and more particularly to communications methods supporting encoding and decoding and related devices and nodes. Background Art

[0002] In communication networks, obtaining good performance and capacity can be challenging given the communication protocols, their parameters, and the physical environment in which the communication network is deployed.

[0003] For example, despite the increasing capacity of telecommunication networks, it still makes sense to limit the resource usage required per user. In mobile telecommunication networks, lower resource usage per call means that the mobile telecommunication network can serve a large number of users in parallel. Lower resource usage also results in lower power consumption in devices on the user side (such as in terminal devices) and in devices on the network side (such as in network nodes). This translates into energy and cost savings for network operators, while in terminal devices one can experience extended battery life and increased talk time.

[0004] One mechanism for reducing the amount of resource usage required for voice communication applications in mobile telecommunication networks is to exploit natural pauses in speech. In more detail, in most conversations only one party is active at a time, so speech pauses in one communication direction will typically take up more than half of the signal. One way to exploit this property to reduce the amount of resource usage required is to employ a discontinuous transmission (DTX) system, in which the active signal encoding is interrupted during speech pauses.

[0005] The encoding process is performed on segments of the audio signal, called frames, where input audio samples during a time interval (typically 10 to 20 ms) are buffered and used by the encoder to extract parameters to be sent to the decoder.

[0006] During speech pauses, so-called SID (Silence Insertion Descriptor) frames are usually sent, which encode the background noise at a very low bit rate to allow a comfort noise generator (CNG) system at the receiving end to fill the pause with background noise that has similar characteristics to the original noise. Compared to keeping silence during speech pauses, CNG makes the sound more natural because the background noise is maintained instead of being switched on and off along with the speech. Complete silence during speech pauses is usually perceived as annoying and often leads to the misinterpretation that the call has been disconnected.

[0007] DTX systems may also rely on a voice activity detector (VAD) which indicates to the transmitting device whether to use active signal encoding or low-rate background noise encoding. In this regard, the transmitting device may be configured to distinguish other source types by using a (generic) sound activity detector (GSAD or SAD) which not only distinguishes speech from background noise, but may also be configured to detect music or other signal types that are considered relevant. Figure 1 A block diagram of a DTX system 100 is shown in FIG.

[0008] exist Figure 1 , input audio is received by VAD 102, speech / audio encoder 104, and CNG encoder 106. VAD 102 indicates whether a "high" bit rate is sent from speech / audio encoder 104 or a "low" bit rate is sent from CNG encoder 106.

[0009] Communication services can be further enhanced by supporting stereo or multi-channel audio transmission. In these cases, the DTX / CNG system can also take into account the spatial characteristics of the signal in order to provide comfort noise that sounds pleasant.

[0010] A common mechanism for generating comfort noise is to send information about the energy and spectral shape of the background noise during speech pauses. This can be done using much fewer bits than conventional encoding of speech segments. Figure 2 As shown, the frequency of sending this information is usually lower than the frequency of sending in the active segment. Figure 2 In , the active segments are shown as activity codes, while the information about the energy and spectral shape of the background noise during speech pauses is shown as CN codes.

[0011] A common feature in DTX systems is to add a so-called "delay period" to the VAD decision, such as Figure 3 During this period, activity coding will still be used, even though the VAD decision indicates that it should not be. This is to avoid short CNG segments in the middle of longer activity segments (e.g., during breathing pauses in speech utterances). Parameters for CNG generation can be estimated during this period.

[0012] On the receiving side, comfort noise is generated by creating a pseudo-random signal and then shaping the spectrum of the signal using a filter based on information received from the transmitting device. The signal generation and spectrum shaping can be performed in the time domain or the frequency domain.

[0013] For stereo operation, additional parameters are sent to the receiving side. In a typical stereo signal, channel pairs show a high degree of similarity or correlation. The most advanced stereo coding schemes exploit this correlation by adopting parametric coding, in which a single channel is encoded with high quality and supplemented with a parametric description, enabling the reconstruction of a complete stereo image. The process of reducing a channel pair to a single channel is often called downmixing, and the resulting channels are called downmixed or downmixed channels. The downmixing process typically attempts to maintain energy by aligning the inter-channel time difference (ITD) and inter-channel phase difference (IPD) before mixing the channels. In order to maintain the energy balance of the input signal, the inter-channel level difference (ILD) is also measured. The ITD, IPD and ILD are then encoded and can be used for the reverse upmixing process when reconstructing the stereo channel pair at the decoder. Figure 4 and Figure 5 A block diagram of a parametric stereo encoder 400 and a decoder 500 is shown.

[0014] exist Figure 4 In the embodiment of the present invention, the time domain stereo input is received by the stereo processing and downmixing module 402. The stereo processing and downmixing module 402 processes the time domain stereo input signal and generates a mono downmix signal and stereo parameters. The mono downmix signal is received by the mono speech / audio encoder 404, which processes the mono downmix signal and generates an encoded mono signal. The encoded mono signal and stereo parameters are sent to a decoder, such as the parameter stereo decoder 500.

[0015] exist Figure 5 In the embodiment, the encoded mono signal is received by a mono voice / audio decoder 502, which decodes the encoded mono signal and generates a mono downmix signal. The mono downmix signal and stereo parameters are received by a stereo processing and upmix decoder 504, which processes the mono downmix signal and stereo parameters and generates a time-domain stereo output. The time-domain stereo output can be stored or sent to an audio player for playback.

[0016] In addition to ITD, IPD, and ILD, the coherence between the left and right channels can also be calculated at the encoder and sent to the receiving side. Coherence basically describes how well the left and right signals are related at different frequencies.

[0017] For DTX operation and CNG, the parametric representation of spatial characteristics (stereo images in the case of stereo audio) is particularly relevant as it is a compact representation. The same or similar parameters as for the parametric stereo coding mode used for active frames can be sent in silence insertion descriptor (SID) frames for comfort noise generation at the decoder. However, for SID frames, larger quantization errors can be allowed without significant perceptual degradation, which means that the number of bits used to represent the spatial characteristics of CNG can be even lower than for the case for active coded frames.

[0018] If coherence parameters are used to represent the characteristics of the spatial audio of the CNG, the coherence can be reconstructed at the decoder and a comfort noise signal with characteristics similar to the original sound can be created. For further details, see U.S. Patent Application Publication No. 20170047072. Note that additional parameters (e.g., ILD, IPD, ITD parameters) are usually required to capture / represent all the most perceptually relevant spatial characteristics, and these additional parameters are sent in the SID frame together with the coherence.

[0019] A solution for efficient representation of coherence is described in PCT publication WO2019193173, where the coherence is calculated at a high frequency resolution in the transmitter, then divided into a small number of bands, and the coherence within each band is weighted and combined into one value for each band. A vector containing the coherence of each band is then encoded and sent to the decoder.

[0020] The stereo encoder receives a channel pair As input, and Represents frames Sample index of The left and right channel input signals are of length. The sampling frequency is , where the length of the frame may include overlap (lookahead samples as well as storage of past samples). Typically, 20ms of new audio samples are buffered and included in the frame being encoded.

[0021] Coding parameters such as ITD are estimated on a per-frame basis at the encoding side and sent to the decoder. If there is no significant gain in using a certain parameter during encoding, it is usually not sent either. In the case of ITD, this would be the case when the left and right signals are more or less uncorrelated.

[0022] The input signal is transformed into the frequency domain by, for example, DFT (Discrete Fourier Transform) or any other suitable filter bank or transform, such as QMF, Hybrid QMF (Quadrature Mirror Filter) or MDCT (Modified Discrete Cosine Transform). In case of using DFT or MDCT, the input signal is usually windowed before the transform. The choice of window depends on various parameters, such as time and frequency resolution characteristics, algorithm delay (overlap length), reconstruction properties, etc. In case of DFT, the spectrum of the left and right audio channels can be obtained as:

[0023]

[0024]

[0025] in, is the selected window function.

[0026] For frequency Vocal tract coherence The general definition is

[0027]

[0028] in, and Indicates two channels and The spectrum of is the cross spectrum. Operating in the DFT domain, the coherence can be estimated based on the cross spectrum and the power spectrum according to the following formula:

[0029]

[0030]

[0031] in, represents complex conjugate.

[0032] However, this relies on good estimates of the cross-spectrum and the power spectrum, which can be obtained, for example, using the well-known Welch method. Another approach for stable coherence estimation is to convert the short-time spectrum , and Before being used in the coherence calculation, it is low-pass filtered using a first-order low-pass filter, as shown in the following equation:

[0033]

[0034]

[0035]

[0036] Then, the coherence can be obtained as:

[0037]

[0038] Rather small values ​​of α are needed to obtain good and stable coherence estimates. Figure 7 Two examples are shown, where the fixed filter coefficients (a) and (b) has been applied to estimate the coherence of two signals with a fixed coherence (0.2 over all frequencies). It can be seen that in the initial frame where the cross-spectrum and power spectrum smoothing filters only contain information from the current frame, the coherence will be 1 for all frequencies in both cases. However, for later frames there is a significant difference. While for (b) the coherence estimate asymptotically approaches the true coherence value, for (a) there is a lot of noise in the estimate. Figure 8 By averaging over frequency, we can see that for (b) the coherence estimate is indeed close to the true coherence (0.2 in this case), while for (a) there is a clear bias in the coherence estimate.

[0039] In a DTX solution where coherence is used only for the generation of comfort noise, the update of the low-pass filter can be skipped during speech segments (i.e., when the VAD indicates speech or active content). The reason is that if this is not done, the speech signal will still be in the low-pass filter state for some time after the speech has stopped and a new comfort noise segment has started, which will lead to a biased coherence estimate of the background signal.

[0040] To reduce the number of bits used to encode the coherence value, the spectrum is divided into frequency bands, such as Figure 6 and the following equation:

[0041]

[0042] in, is a vector containing the boundaries between frequency bands.

[0043] The width of these frequency bands is intended to simulate the frequency resolution of human auditory perception, with narrow bands used for low frequencies and progressively larger bandwidths used for higher frequencies.

[0044] Instead of using the average coherence within a frequency band, a weighted mean may be used for each frequency band, where a mono signal (e.g., ) is used as a weighting function. Details can be found in PCT publication WO2019193156. Using the weighting function, the equation can be written as

[0045] Summary of the invention

[0046] There are some challenges at present. If the smoothed left and right power spectra and cross-spectrum used in the coherence calculation contain the time part where the speech signal exists, they may not reflect the characteristics of the background noise and cause the incorrect generation of comfort noise. One reason for this may be that the last frame before the speech segment contains the beginning of the speech segment. The energy of this part may be too low and / or other features of the audio signal may not be enough to trigger the VAD to detect speech, but it may still have a negative impact on the background noise coherence estimation.

[0047] One solution to this problem is to store the left and right spectra and the cross spectrum of the previous frame and remove it from the low pass filter state if the next frame is a speech frame. However, in the case of a high resolution DFT this may mean having to spend several thousand words of memory to store the previous values.

[0048] Certain aspects of the present disclosure and embodiments thereof may provide solutions to these or other challenges and improve coherence estimation in speech pauses by minimizing the impact of speech portions. Various embodiments described herein determine coherence for a small number of frequency bands and use the coherence when creating filter states for a low pass filter that reflect previous background noise.

[0049] According to some embodiments, a method in an encoder that enables the use of estimated coherence parameters to generate comfort noise in a network using discontinuous transmission (DTX) includes: receiving a time domain audio input including an audio input signal. The method includes processing the audio input signal frame by frame by: encoding the active content of each audio input signal at a first bit rate until an inactive period is detected in the audio input signal; switching the encoding from the active encoded content to the inactive encoding during the inactive period so as to encode background noise at a second bit rate; estimating the coherence parameters during the inactive period based on low-pass filtering of a cross-spectrum or an average of the cross-spectrum, wherein estimating the coherence parameters includes reinitializing the low-pass filter state of the cross-spectrum based on the coherence parameters from the previous inactive period; encoding the estimated coherence parameters; and initiating sending the encoded active content, the encoded background noise, and the encoded coherence parameters to a decoder.

[0050] Similar encoders, computer programs and computer program products are also provided.

[0051] Certain embodiments may provide one or more of the following technical advantages. Various embodiments make comfort noise sound more natural and avoid annoying effects of sudden changes in spatial characteristics during CNG after changing from activity coding to CNG. Specifically, it is avoided that DTX starts with a period of comfort noise colored by activity content and then suddenly changes to comfort noise closer to the original input noise after a certain period of time. Various embodiments may estimate coherence and minimize the influence of speech parts in the estimation. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] The accompanying drawings illustrate certain non-limiting embodiments of the inventive concept, and are included to provide a further understanding of the present disclosure, and are incorporated into and constitute a part of this application. In the drawings:

[0053] Figure 1 is a block diagram of the DTX system;

[0054] Figure 2 is a flow chart showing CNG parameter encoding and transmission;

[0055] Figure 3 is a flow chart showing a VAD (or DTX) hold-off period;

[0056] Figure 4 is a block diagram of a parametric stereo encoder;

[0057] Figure 5 is a block diagram of a parametric stereo decoder according to some embodiments;

[0058] Figure 6 is a diagram of coherent frequency band segmentation according to some embodiments;

[0059] Figure 7 is a graphical representation of a coherence estimate of a stereo signal according to some embodiments;

[0060] Figure 8 is a graph showing the average over frequency of a coherence estimate of a stereo signal with fixed filter coefficients;

[0061] Fig. 9 is a graph showing estimated coherence compared to reset cross-spectrum and power spectrum according to various embodiments;

[0062] Fig.10 is a flow chart illustrating filter state initialization according to some embodiments;

[0063] Fig.11 is a flow chart illustrating a coherence memory shift according to some embodiments;

[0064] Fig.12is a block diagram illustrating an operating environment of a parametric stereo encoder and a parametric stereo decoder according to some embodiments;

[0065] Fig.13 is a block diagram of an encoder according to some embodiments;

[0066] Fig.14 is a block diagram of a decoder according to some embodiments;

[0067] Fig.15 is a block diagram of a host according to some embodiments;

[0068] Fig.16 is a block diagram of a virtualization environment according to some embodiments; and

[0069] Figures 17 to 24 is a flow chart illustrating the operation of an encoder according to some embodiments. DETAILED DESCRIPTION

[0070] Some embodiments contemplated herein will now be described more fully with reference to the accompanying drawings. Embodiments are provided by way of example to convey the scope of subject matter to those skilled in the art, wherein examples of embodiments of the inventive concept are shown. However, the inventive concept may be embodied in a variety of different forms and should not be construed as being limited to the embodiments set forth herein. On the contrary, these embodiments are provided so that the disclosure will be comprehensive and complete, and the scope of the inventive concept is fully conveyed to those skilled in the art. It should also be noted that these embodiments are not mutually exclusive. A component from one embodiment may be assumed by default to be present in / used in another embodiment.

[0071] As previously noted, avoiding the annoying effects of a sudden change in spatial characteristics during CNG after changing from active encoding to CNG will make the comfort noise sound more natural.

[0072] Various embodiments calculate a set of coherence values ​​for each frame that VAD or SAD signals non-speech. These coherence values ​​are stored at least back in time for two frames. Fig.10 The situation of tracing back two frames in time is shown. In some embodiments, more coherence values ​​can be used. In the following description, the last two frames will be used to describe these embodiments.

[0073] When a new inactive segment begins, the cross-spectrum The low-pass filter state is initialized to

[0074] ,

[0075]

[0076] is a complex number with absolute value = 1 and random phase, where It is the frequency band The frequency coefficient set of .

[0077] Note that the smoothed left spectrum and right spectrum is used as follows:

[0078]

[0079]

[0080] in, is the smoothing coefficient.

[0081] This initialization will cause the correlation calculation to give This is not an important thing in itself. Can be used directly for the first frame in an inactive segment instead of using the updated filter state to recalculate the correlation. Importantly, The filter state starts at a point that gives the same coherence as the end of the previous inactive segment.

[0082] Note that in other embodiments, other frames near the end of the last inactive period may be used. For example, , which will most likely give very little performance difference and only a small increase in memory usage.

[0083] In some scenarios, has been set to zero at the beginning of the VAD holdoff period and then updated during the holdoff period and the first inactive frame. will not have been updated enough times to give a reliable coherence estimate, but the initialization value is calculated using the The phase information of has been shown to give an improvement over the case where random numbers are used. This is achieved by Scaling it by its absolute value gives an absolute value of 1 but a phase of This is done by taking the complex number of the phase.

[0084]

[0085]

[0086] in, It is the frequency band The frequency coefficient set of .

[0087] For clarity, use To calculate the first , and the result is calculated by normalization And further scale it To maintain the phase of the calculation results.

[0088] The special case that needs to be handled is when the inactive segment is only one frame long. Will be used to initialize filter state, as described above. At this point, The last frame of the previous inactive segment will be taken, i.e., the frame that may contain the beginning of speech. will be updated to , but in this case this will result in This may include the start frame which one would like to avoid. The solution to this problem is to not update the first frame of an inactive segment Instead, the update is first performed in the second frame of the inactive segment.

[0089] The problem is Fig.11 It is shown in Fig.11 In , the dashed frame is used for an inactive segment of one frame length, and the solid frame containing the start will be used for the next inactive segment. If no update is done in the first frame of an inactive segment, a dashed frame will be used instead.

[0090] Fig. 9 The advantage of the disclosed method in estimating the coherence of inactive coding segments, which is sent in SID frames for CNG at the receiving side, is shown. Figure 8 Similarly, the average over frequency is plotted. The true coherence of the signal is fixed to 0.2 at all frequencies. It can be seen that the proposed method maintains a good estimate of the coherence, while resetting the cross spectrum and power spectrum restarts the estimation process. Restarting the estimation process means that there will be an inaccurate coherence estimate at the beginning of the second inactive segment.

[0091] One reason to reset the cross spectrum during active segments may be to improve the ITD estimate for inactive segments, which would otherwise rely on cross spectrum data that is heavily influenced by the ITD of active speech segments. It can also be beneficial, and from Figure 7It can be seen that having larger filter coefficients will lead to unstable coherence estimates. However, with the proposed method of reinitializing the cross-spectrum, stable and reliable coherence estimates as well as accurate ITD estimates can be obtained while minimizing the memory footprint. The same memory can be used to store the cross-spectrum filter state (which can be reset during the delay period and updated faster to be subsequently copied to another cross-spectrum filter state and used for ITD estimation during the inactive coding (CNG) segment), and to store the cross-spectrum estimate for estimating coherence in the inactive coding segment. Therefore, two filter state vectors can be used for cross-spectrum estimation instead of three filter state vectors, which can mean a lot of memory for high-resolution DFT, especially for applications in codecs targeted at mobile devices or any other device with limited memory capacity.

[0092] Before describing the operation from the perspective of encoder 400, Fig.12 1 is a block diagram of an example of an operating environment 1200 in which the encoder 400 and the decoder 500 can be implemented. Fig.12 In the embodiment of the present invention, the encoder 400 receives data (e.g., audio files) to be encoded from an entity (e.g., host 1204) and / or from a storage device 1206 via a network 1202. In some embodiments, the host 1204 can communicate directly with the encoder 400. The 1204 hosts can be included in various combinations of hardware and / or software, including UE, mobile phones, terminals, independent servers, blade servers, servers implemented in the cloud, distributed servers, virtual machines, containers, or processing resources in a server group. The encoder 400 encodes the audio file as described herein, and stores the encoded audio file in a storage device 1206 and / or sends the encoded audio file to the decoder 500 via a network 1208. The decoder 500 decodes the audio file, and sends the decoded audio file to an audio player such as a multi-channel audio player 1210 for playback. The decoder 500 can be in a UE, a mobile phone, a terminal, etc. The multi-channel audio player 1210 can be included in a user device, a terminal, a mobile phone, etc.

[0093] Fig.13is a block diagram illustrating units of an encoder 400 configured to encode audio frames according to various embodiments herein. As shown, the encoder 400 may include a network interface circuit 1305 (also referred to as a network interface) configured to provide communication with other devices / entities / functions / etc. The encoder 400 may also include a processing circuit 1301 (also referred to as a processor and processor circuit) coupled to the network interface circuit 1305 and a memory circuit 1303 (also referred to as a memory) coupled to the processing circuit. The memory circuit 1303 may include computer-readable program code that, when executed by the processing circuit 1301, causes the processing circuit to perform operations according to embodiments disclosed herein.

[0094] According to other embodiments, the processing circuit 1301 can be defined to include a memory, so that a separate memory circuit is not required. As discussed herein, the operation of the encoder 400 can be performed by the processing circuit 1301 and / or the network interface 1305. For example, the processing circuit 1301 can control the network interface 1405 to send communications to the decoder 500 and / or receive communications from one or more other network nodes / entities / servers (e.g., other encoder nodes, library servers, etc.) through the network interface 1305. In addition, modules can be stored in the memory 1303, and these modules can provide instructions so that when the instructions of the modules are executed by the processing circuit 1301, the processing circuit 1301 performs the corresponding operations.

[0095] Fig.14 1 is a block diagram illustrating units of a decoder 500 configured to decode audio frames according to some embodiments of the present inventive concept. As shown, the decoder 120 may include a network interface circuit 1405 (also referred to as a network interface) configured to provide communication with other devices / entities / functions / etc. The decoder 500 may also include a processing circuit 1401 (also referred to as a processor or processor circuit) coupled to the network interface circuit 1405 and a memory circuit 1403 (also referred to as a memory) coupled to the processing circuit. The memory circuit 1403 may include computer-readable program code that, when executed by the processing circuit 1401, causes the processing circuit to perform operations according to the embodiments disclosed herein.

[0096] According to other embodiments, the processing circuit 1401 may be defined to include a memory, thereby eliminating the need for a separate memory circuit. As discussed herein, the operations of the decoder 500 may be performed by the processor 1401 and / or the network interface 1405. For example, the processing circuit 1401 may control the network interface circuit 1405 to receive communications from the encoder 400. In addition, modules may be stored in the memory 1403, and these modules may provide instructions such that when the instructions of the modules are executed by the processing circuit 1401, the processing circuit 1401 performs the corresponding operations.

[0097] Fig.15 1 is a block diagram illustrating elements of a host 1204 configured to provide an audio file to an encoder to encode the audio file and send the encoded audio file to a decoder 500 according to some embodiments. As shown, the host 1204 may include a network interface circuit 1505 (also referred to as a network interface) configured to provide communication with other devices / entities / functions, etc. The host 1204 may also include a processing circuit 1501 (also referred to as a processor or processor circuit) coupled to the network interface circuit 1505, and a memory circuit 1503 (also referred to as a memory) coupled to the processing circuit. The memory circuit 1503 may include computer-readable program code that, when executed by the processing circuit 1501, causes the processing circuit to perform operations according to embodiments disclosed herein.

[0098] According to other embodiments, processing circuit 1501 may be defined to include memory, thereby eliminating the need for a separate memory circuit. As discussed herein, operations of host 1204 may be performed by processor 1501 and / or network interface 1505. For example, processing circuit 1501 may control network interface circuit 1505 to send communications to encoder 400. In addition, modules may be stored in memory 1503, and these modules may provide instructions such that when the instructions of the modules are executed by processing circuit 1501, processing circuit 1501 performs corresponding operations.

[0099] In some embodiments, the encoder 400 and the decoder 500 may be virtualized by distributing the encoder 400 and / or the decoder 500 across various components. Fig.161600 is a block diagram showing an example of a virtualized environment 1600, in which the functions implemented by some embodiments can be virtualized. In this context, virtualization means creating a virtual version of an apparatus or device that can include a virtualized hardware platform, storage device, and network resources. As used herein, virtualization can be applied to any device or component thereof described herein, and involves an implementation in which at least a portion of the functions are implemented as one or more virtual components. Some or all of the functions described herein can be implemented as virtual components performed by one or more virtual machines (VMs), which are implemented in one or more virtual environments 1600 hosted by one or more hardware nodes (e.g., hardware computing devices operating as network nodes, UEs, core network nodes, or hosts). In addition, in embodiments where the virtual nodes do not require a radio connection (e.g., a core network node or a host), the nodes can be fully virtualized.

[0100] An application 1602 (which may alternatively be referred to as a software instance, a virtual application, a network function, a virtual node, a virtual network function, etc.) runs in a virtualized environment 1600 to implement some of the features, functions, and / or benefits of some embodiments disclosed herein.

[0101] Hardware 1604 includes processing circuitry, memory storing software and / or instructions that can be executed by the hardware processing circuitry, and / or other hardware devices described herein (such as network interfaces, input / output interfaces, etc.). The software can be executed by the processing circuitry to instantiate one or more virtualization layers 1606 (also referred to as hypervisors or virtual machine monitors (VMMs)), provide VMs 1608A and VMs 1608B (one or more of which can be generally referred to as VMs 1608), and / or perform any functions, features, and / or benefits described in connection with some embodiments described herein. Virtualization layer 1606 can present a virtual operating platform to VMs 1608 that looks like network hardware.

[0102] VM 1608 includes virtual processing, virtual memory, virtual network or interface, and virtual storage, and can be run by corresponding virtualization layer 1606. Different embodiments of instances of virtual device 1602 can be implemented on one or more VMs 1608, and these implementations can be made in different ways. In some contexts, virtualization of hardware is referred to as network function virtualization (NFV). NFV can be used to unify numerous network device types onto industry standard high-capacity server hardware, physical switches, and physical storage devices (which can be located in data centers and customer premises equipment (CPE)).

[0103] In the context of NFV, VM 1608 can be a software implementation of a physical machine that runs programs as if they were executed on a physical, non-virtualized machine. Each VM 1608 and the hardware portion of hardware 1604 that executes the VM (whether it is hardware dedicated to the VM and / or hardware shared by the VM with other VMs) form a separate virtual network unit. Still in the context of NFV, a virtual network function is responsible for handling specific network functions running in one or more VMs 1608 on hardware 1604 and corresponding to an application 1602.

[0104] Hardware 1604 may be implemented in a standalone network node with general or specific components. Hardware 1604 may implement some functions via virtualization. Alternatively, hardware 1604 may be part of a larger hardware cluster (e.g., in a data center or CPE) in which many hardware nodes work together and are managed by management and coordination 1610, which in particular oversees the lifecycle management of application 1602. In some embodiments, hardware 1604 is coupled to one or more radio units, each of which includes one or more transmitters and one or more receivers that may be coupled to one or more antennas. The radio unit may communicate directly with other hardware nodes via one or more appropriate network interfaces, and may be used in conjunction with virtual components to provide a virtual node (e.g., a radio access node or base station) with radio capabilities. In some embodiments, some signaling may be provided by using a control system 1612, which may alternatively be used for communication between hardware nodes and radio units.

[0105] Reference will now be made to some embodiments according to the inventive concept Fig.17 The flow chart of the encoder 400 (using Figure 4 and Fig.13 For example, a module can be stored in Fig.13 The modules are stored in the memory 1303, and these modules can provide instructions so that when the instructions of the modules are executed by the corresponding encoder processing circuit 1301, the encoder 400 performs the corresponding operations of the flowchart.

[0106] Fig.17 The operations performed by the encoder 400 in various embodiments are shown. Fig.17 In block 1701, the encoder 400 receives a time domain audio input including an audio input signal. The audio input signal may be speech, music, or a combination thereof.

[0107] In block 1703, the encoder 400 processes the audio input signal frame by frame, as shown in blocks 1705 to 1711. The encoder 400 may perform this processing in the time domain or the frequency domain.

[0108] In blocks 1705 to 1711, the encoder 400 encodes each audio input signal. Specifically, in block 1705, the encoder 400 encodes the active content of each audio input signal at a first bit rate until an inactive period is detected in the audio input signal. A VAD (e.g., VAD 102) or a SAD may be used to detect the inactive period, as described above.

[0109] In block 1707, during the pause period, the encoder 400 switches encoding from the active encoding content to the inactive encoding to encode the background noise at a second bit rate. The second bit rate is typically less than the first bit rate, as described above.

[0110] In block 1709 , the encoder 400 estimates coherence parameters during the inactive period based on low pass filtering of the cross spectrum or averaging of the cross spectrum, wherein estimating the coherence parameters includes initializing a low pass filter state of the cross spectrum based on coherence parameters from a previous inactive period.

[0111] In such Fig.18 In some embodiments shown, when estimating the coherence parameters, in block 1801, the encoder 400 reinitializes the first cross-spectral low-pass filter based on the coherence parameters from the previous inactive coding period in the first coding frame after the active coding. status.

[0112] In some embodiments, the encoder 400 reinitializes the first cross-spectral low-pass filter based on the last two frames from the previous inactive encoding period. status.

[0113] In other embodiments, the coherence parameter may be various functions of previous coherence values. For example, encoder 400 may estimate the coherence parameter by picking the second to last coherence value in a previous inactive period, taking an average of the last few estimated coherence parameters (and possibly excluding the last coherence parameter), taking a weighted average of previous coherence values, creating a filtered version of an earlier coherence value, e.g., and use Rather than To reinitialize wait.

[0114] exist Fig.19 In block 1901 of the embodiment, the encoder 400 starts the second low pass filter during the DTX delay period. Updates.

[0115] In block 1711 , the encoder 400 encodes the estimated coherence parameters.

[0116] In block 1713 , the encoder 400 initiates sending the encoded active content, the encoded background noise, and the coherence parameters to the decoder 500 .

[0117] In some embodiments where the audio input signals are processed frame by frame, the encoder 400 processes the audio input signals frame by frame to generate a mono mix signal, and the encoder 400 encodes the active content of each audio input signal by encoding the active content of the mono mix signal.

[0118] In other embodiments, the encoder 400 processes the audio input signal frame by frame to generate a mono downmix signal and one or more stereo parameters. In these embodiments, the encoder 400 encodes the motion content of the mono downmix signal and the one or more stereo parameters.

[0119] In some embodiments, the encoder 400 determines according to the following formula :

[0120]

[0121]

[0122]

[0123]

[0124]

[0125]

[0126] in, represents multiplication, is the low-pass coefficient, It is the frequency band The frequency coefficient set of is a vector containing the limits between the frequency bands, and is a complex number with absolute value = 1 and random phase.

[0127] In some other embodiments, the encoder 400 determines according to the following formula :

[0128]

[0129]

[0130]

[0131]

[0132]

[0133]

[0134] in, represents multiplication, is the low-pass coefficient, It is the frequency band The frequency coefficient set of is a vector containing the limits between the frequency bands. As before, for clarity, we use To calculate the first , and the result is calculated by normalization And further scale it up To maintain the phase of the calculation results.

[0135] The above is used to determine In this way, the encoder 400 is as follows Fig. 20 As shown in the box 2001, the weighted function is used to For example, as described above, the encoder 400 may use a weighting function according to the following formula: To perform weighting:

[0136]

[0137] in, is the discrete Fourier transform (DFT) energy spectrum of the mono signal which is a downmix of the audio input signal.

[0138] In some embodiments, the previous inactive period may consist of only one frame. In this case, The processing may cause the starting frame to be part of the comfort noise, which is undesirable. To solve this problem, Fig.21 As shown in block 2101 of FIG. 1 , the encoder 400 does not update the first frame having an inactive period of multiple frames. Instead, it is updated in the second frame with a multi-frame inactive period .

[0139] In other embodiments, a dedicated cross-correlation estimate may be used. Fig. 22 As shown in block 2201 of , the encoder 400 performs a dedicated cross-correlation estimation, which is updated for the cross-spectrum only during the pause period and / or during the DTX delay frame, and uses the dedicated cross-correlation estimation for coherence estimation in the inactive period.

[0140] In such Fig.23In another embodiment shown in block 2301 of , the encoder 400 accelerates the smoothing of the cross spectrum by low-pass filtering by resetting the cross-spectrum low-pass filter state before any update in the DTX delay period or before any update in the pause period. Additionally or alternatively, the filter coefficients may be increased to speed up the effect of new frames being processed.

[0141] In such Fig.23 In another embodiment shown in block 2301 of , the encoder 400 accelerates the smoothing of the cross spectrum by low-pass filtering by replacing the low-pass filter state at the beginning of a delay period or at the beginning of an inactive period.

[0142] In such Fig.24 In another embodiment shown in block 2401 of , the encoder 400 reinitializes the low pass filter state at the beginning of a delay period or at the beginning of an inactive period.

[0143] Although the computing devices described herein (e.g., encoders, decoders, hosts) may include combinations of the hardware components shown, other embodiments may include computing devices with different combinations of components. It should be understood that these computing devices may include any suitable combination of hardware and / or software required to perform the tasks, features, functions, and methods disclosed herein. The determination, calculation, acquisition, or similar operations described herein may be performed by a processing circuit, which may process information by, for example, converting the obtained information to other information, comparing the obtained information or the converted information with information stored in a network node, and / or performing one or more operations based on the obtained information or the converted information, and making a determination based on the results of the processing. In addition, although the components are depicted as a single box located within a larger box or nested within multiple boxes, in reality, the computing device may include multiple different physical components that constitute a single illustrated component, and functions may be divided between different components. For example, a communication interface may be configured to include any component described herein, and / or the functions of a component may be divided between a processing circuit and a communication interface. In another example, the non-computationally intensive functions of any such component may be implemented in software or firmware, and the computationally intensive functions may be implemented in hardware.

[0144] In some embodiments, some or all of the functions described herein may be provided by a processing circuit that executes instructions stored in a memory, which in some embodiments may be a computer program product in the form of a non-transitory computer-readable storage medium. In alternative embodiments, some or all of the functions may be provided by a processing circuit, for example, in a hardwired manner, without executing instructions stored on a separate or discrete device-readable storage medium. In any of these specific embodiments, the processing circuit may be configured to perform the described functions, regardless of whether instructions stored on a non-transitory computer-readable storage medium are executed. The benefits provided by such functionality are not limited to separate processing circuits or other components of a computing device, but are enjoyed by the computing device as a whole and / or are generally enjoyed by end users and wireless networks.

[0145] Example

[0146] Embodiment 1. A method in an encoder (400) for enabling the use of estimated coherence parameters to generate comfort noise in a network using discontinuous transmission (DTX), the method comprising:

[0147] Receiving (1701) a time domain audio input comprising an audio input signal;

[0148] The audio input signal is processed (1703) frame by frame in the following manner:

[0149] encoding (1705) the active content of each audio input signal at a first bit rate until a period of inactivity is detected in the audio input signal;

[0150] switching (1707) encoding from the active encoding content to the inactive encoding during the inactive period to encode the background noise at a second bit rate;

[0151] estimating (1709) a coherence parameter during the inactive period based on low pass filtering of the cross spectrum or averaging of the cross spectrum, wherein estimating the coherence parameter comprises reinitializing a low pass filter state of the cross spectrum based on coherence parameters from a previous inactive period;

[0152] encoding the estimated coherence parameters (1711); and

[0153] Sending (1713) the encoded active content, the encoded background noise, and the encoded coherence parameters to the decoder (500) is initiated.

[0154] Embodiment 2. The method according to embodiment 1, wherein estimating the coherence parameter comprises:

[0155] In a first coding frame after active coding, a first cross-spectral low-pass filter is reinitialized (1801) based on coherence parameters from a previous inactive coding period. status.

[0156] Embodiment 3. The method of embodiment 2, wherein the first cross-spectral low-pass filter is reinitialized based on coherence parameters from a previous inactive encoding period. The state of includes: reinitializing the first cross-spectral low-pass filter based on the last two frames from the previous inactive encoding period status.

[0157] Embodiment 4. The method according to any one of embodiments 2 to 3 further includes:

[0158] The low pass filter starts (1901) during the DTX hold-off period. Updates.

[0159] Embodiment 5. A method according to any one of embodiments 1 to 4, wherein processing the audio input signal frame by frame includes: processing the audio input signal frame by frame to generate a mono downmix signal, and encoding the active content of each audio input signal includes: encoding the active content of the mono downmix signal.

[0160] Embodiment 6. A method according to embodiment 5, wherein processing the audio input signal frame by frame to generate a mono downmix signal includes: processing the audio input signal frame by frame to generate a mono downmix signal and one or more stereo parameters, and encoding the active content of the mono downmix signal includes: encoding the active content of the mono downmix signal and one or more stereo parameters.

[0161] Embodiment 7. The method according to any one of embodiments 2 to 6, wherein: Determine according to the following formula:

[0162]

[0163]

[0164]

[0165]

[0166]

[0167]

[0168] in, represents multiplication, is the low-pass coefficient, It is the frequency band The frequency coefficient set of is a vector containing the limits between the frequency bands, and is a complex number with absolute value = 1 and random phase.

[0169] Embodiment 8. The method according to embodiment 7 further comprises using a weighting function to Weighted (2001).

[0170] Embodiment 9. The method according to embodiment 8, wherein a weighting function is used to The weighting is performed according to the following formula:

[0171]

[0172] in, is the discrete Fourier transform (DFT) energy spectrum of the mono signal which is a downmix of the audio input signal.

[0173] Embodiment 10. The method according to any one of embodiments 2 to 6, wherein: Determine according to the following formula:

[0174]

[0175]

[0176]

[0177]

[0178]

[0179]

[0180] in, represents multiplication, is the low-pass coefficient, It is the frequency band The frequency coefficient set of is a vector containing the boundaries between frequency bands.

[0181] Embodiment 11. The method according to embodiment 10 further comprises using a weighting function to Weighted (2001).

[0182] Embodiment 12. The method according to embodiment 11, wherein a weighting function is used to The weighting is performed according to the following formula:

[0183]

[0184] in, is the discrete Fourier transform (DFT) energy spectrum of the mono signal which is a downmix of the audio input signal.

[0185] Embodiment 13. The method according to any one of embodiments 1 to 12, further comprising:

[0186] Not updated in the first frame of an inactive period of multiple frames , but updated in the second frame with an inactive period of multiple frames (2101).

[0187] Embodiment 14. The method according to any one of embodiments 1 to 12, further comprising:

[0188] A dedicated cross-correlation estimation is performed (2201), which is updated for the cross-spectrum only during the inactive period and / or during the DTX holdover frame, and is used for coherence estimation in the inactive period.

[0189] Embodiment 15. The method according to any one of embodiments 1 to 14, further comprising:

[0190] The cross-spectral low-pass filter state is reset (2301) prior to any update in the DTX holdoff period or prior to any update in the inactive period.

[0191] Embodiment 16. The method according to any one of embodiments 1 to 15, further comprising:

[0192] The low pass filter state is reinitialized (2401) at the beginning of a hold-off period or at the beginning of an inactivity period.

[0193] Embodiment 17. An encoder (400) adapted to enable the use of estimated coherence parameters to generate comfort noise in a network using discontinuous transmission (DTX), the encoder being adapted to:

[0194] Receiving (1701) a time domain audio input comprising an audio input signal;

[0195] The audio input signal is processed (1703) frame by frame in the following manner:

[0196] encoding (1705) the active content of each audio input signal at a first bit rate until a period of inactivity is detected in the audio input signal;

[0197] switching (1707) encoding from the active encoding content to the inactive encoding during the inactive period to encode the background noise at a second bit rate;

[0198] estimating (1709) a coherence parameter during the inactive period based on low pass filtering of the cross spectrum or averaging of the cross spectrum, wherein estimating the coherence parameter comprises: initializing a low pass filter state of the cross spectrum based on coherence parameters from a previous inactive period;

[0199] encoding the estimated coherence parameters (1711); and

[0200] Transmission (1713) of the encoded active content, the encoded background noise, and the encoded coherence parameters to a decoder (500) is initiated.

[0201] Embodiment 18. The encoder (400) according to embodiment 17, wherein the encoder is further adapted to perform the method according to any one of embodiments 2 to 16.

[0202] Embodiment 19. An encoder (400) adapted to enable the use of estimated coherence parameters to generate comfort noise in a network using discontinuous transmission (DTX), the encoder comprising:

[0203] Processing circuit (1301); and

[0204] A memory (1303) is coupled to the processing circuit, wherein the memory includes instructions that, when executed by the processing circuit, cause the encoder to perform operations, the operations comprising:

[0205] Receiving (1701) a time domain audio input comprising an audio input signal;

[0206] The audio input signal is processed (1703) frame by frame in the following manner:

[0207] encoding (1705) the active content of each audio input signal at a first bit rate until a period of inactivity is detected in the audio input signal;

[0208] switching (1707) encoding from the active encoding content to the inactive encoding during the inactive period to encode the background noise at a second bit rate;

[0209] estimating (1709) a coherence parameter during the inactive period based on low pass filtering of the cross spectrum or averaging of the cross spectrum, wherein estimating the coherence parameter comprises: initializing a low pass filter state of the cross spectrum based on coherence parameters from a previous inactive period;

[0210] encoding the estimated coherence parameters (1711); and

[0211] Initiating (1713) sending the encoded activity content, the encoded background noise, and the encoded coherence parameters to the decoder (500).

[0212] Embodiment 20. The encoder (400) of embodiment 19, wherein the memory comprises further instructions which, when executed by the processing circuit, cause the encoder to perform the method of any one of embodiments 2 to 16.

[0213] Embodiment 21. A computer program comprising program code to be executed by a processing circuit (1301) of an encoder (400), whereby execution of the program code causes the encoder (400) to perform operations comprising:

[0214] Receiving (1701) a time domain audio input comprising an audio input signal;

[0215] The audio input signal is processed (1703) frame by frame in the following manner:

[0216] encoding (1705) the active content of each audio input signal at a first bit rate until a period of inactivity is detected in the audio input signal;

[0217] switching (1707) encoding from the active encoding content to the inactive encoding during the inactive period to encode the background noise at a second bit rate;

[0218] estimating (1709) a coherence parameter during the inactive period based on low pass filtering of the cross spectrum or averaging of the cross spectrum, wherein estimating the coherence parameter comprises: initializing a low pass filter state of the cross spectrum based on coherence parameters from a previous inactive period;

[0219] encoding the estimated coherence parameters (1711); and

[0220] Initiating (1713) sending the encoded activity content, the encoded background noise, and the encoded coherence parameters to the decoder (500).

[0221] Embodiment 22. The computer program according to embodiment 21 comprises further program code to be executed by a processing circuit of an encoder, whereby execution of the further program code causes the encoder (400) to perform operations according to any one of embodiments 2 to 16.

[0222] Embodiment 23. A computer program product comprising a non-transitory storage medium comprising program code to be executed by a processing circuit (1301) of an encoder (400), whereby execution of the program code causes the encoder (400) to perform operations comprising:

[0223] Receiving (1701) a time domain audio input comprising an audio input signal;

[0224] The audio input signal is processed (1703) frame by frame in the following manner:

[0225] encoding (1705) the active content of each audio input signal at a first bit rate until a period of inactivity is detected in the audio input signal;

[0226] switching (1707) encoding from the active encoding content to the inactive encoding during the inactive period to encode the background noise at a second bit rate;

[0227] estimating (1709) a coherence parameter during the inactive period based on low pass filtering of the cross spectrum or averaging of the cross spectrum, wherein estimating the coherence parameter comprises: initializing a low pass filter state of the cross spectrum based on coherence parameters from a previous inactive period;

[0228] encoding the estimated coherence parameters (1711); and

[0229] Initiating (1713) sending the encoded activity content, the encoded background noise, and the encoded coherence parameters to the decoder (500).

[0230] Embodiment 24. A computer program product according to embodiment 23, wherein the non-temporary storage medium comprises further program code to be executed by the processing circuit (1301) of the encoder (400), whereby execution of the further program code causes the encoder (400) to perform the operations described in any one of embodiments 2 to 16.

[0231] Embodiment 25. A method implemented by a host (1204) configured to operate in a communication system, the communication system also including an encoder, a decoder, and an audio player, the method comprising:

[0232] providing user data to a decoder (500) for decoding an audio file for an audio player; and

[0233] A transmission is initiated to an audio player via a cellular network including an encoder (400), the transmission carrying an audio file, wherein the encoder (400) performs the following operations to send user data from a host to a decoder (500):

[0234] Receiving (1701) a time domain audio input comprising an audio input signal;

[0235] The audio input signal is processed (1703) frame by frame in the following manner:

[0236] encoding (1705) the active content of each audio input signal at a first bit rate until a period of inactivity is detected in the audio input signal;

[0237] switching (1707) encoding from the active encoding content to the inactive encoding during the inactive period to encode the background noise at a second bit rate;

[0238] estimating (1709) a coherence parameter during the inactive period based on low pass filtering of the cross spectrum or averaging of the cross spectrum, wherein estimating the coherence parameter comprises: initializing a low pass filter state of the cross spectrum based on coherence parameters from a previous inactive period;

[0239] encoding the estimated coherence parameters (1711); and

[0240] Initiating (1713) sending the encoded activity content, the encoded background noise, and the encoded coherence parameters to the decoder (500).

[0241] Embodiment 26. The method according to embodiment 25, wherein the encoder (400) is further configured to:

[0242] Perform the method according to any one of embodiments 2 to 16.

[0243] Embodiment 27. A host (1204) configured to operate in a communication system to provide an over-the-top OTT service, the host comprising:

[0244] A processing circuit (1501) configured to provide an audio file; and

[0245] A network interface (1505) configured to initiate transmission of an audio file to an encoder (400) in a cellular network to send it to a decoder (500), the encoder (400) having a network interface (1305) and a processing circuit (1301), the processing circuit (1301) of the encoder (400) being configured to perform the following operations to send the audio file from the host to the decoder (500):

[0246] Receiving (1701) a time domain audio input comprising an audio input signal;

[0247] The audio input signal is processed (1703) frame by frame in the following manner:

[0248] encoding (1705) the active content of each audio input signal at a first bit rate until a period of inactivity is detected in the audio input signal;

[0249] switching (1707) encoding from the active encoding content to the inactive encoding during the inactive period to encode the background noise at a second bit rate;

[0250] estimating (1709) a coherence parameter during the inactive period based on low pass filtering of the cross spectrum or averaging of the cross spectrum, wherein estimating the coherence parameter comprises: initializing a low pass filter state of the cross spectrum based on coherence parameters from a previous inactive period;

[0251] encoding the estimated coherence parameters (1711); and

[0252] Initiating (1713) sending the encoded activity content, the encoded background noise, and the encoded coherence parameters to the decoder (500).

[0253] Embodiment 28. The host according to embodiment 27, wherein the processing circuit of the encoder (400) is further configured to:

[0254] Perform the method according to any one of embodiments 2 to 16.

[0255] References are as follows:

[0256] U.S. Patent Application Publication No. 20200194013

[0257] U.S. Patent No. 11,417,348

[0258] U.S. Patent No. 11,404,069.

Claims

1. A method in an encoder (400) for enabling the use of estimated coherence parameters to generate comfort noise in a network using discontinuous transmission (DTX), the method comprising: Receiving (1701) a time domain audio input comprising an audio input signal; The audio input signal is processed (1703) frame by frame by: encoding (1705) the active content of each audio input signal at a first bit rate until a period of inactivity is detected in the audio input signal; switching (1707) encoding from the active encoding content to the inactive encoding during the inactive period to encode the background noise at a second bit rate; estimating (1709) a coherence parameter during the inactive period based on low pass filtering of the cross spectrum or averaging of the cross spectrum, wherein estimating the coherence parameter comprises: reinitializing a low pass filter state of the cross spectrum based on coherence parameters from a previous inactive period; and The estimated coherence parameters are encoded (1711).

2. The method according to claim 1, further comprising: Sending (1713) the encoded active content, the encoded background noise, and the encoded coherence parameters to the decoder (500) is initiated.

3. The method according to any one of claims 1 to 2, wherein: Estimating the coherence parameter comprises: In a first coding frame after active coding, a first cross-spectral low-pass filter is reinitialized (1801) based on coherence parameters from a previous inactive coding period. status.

4. The method according to claim 3, wherein: The first cross-spectral low-pass filter is reinitialized based on coherence parameters from a previous inactive encoding period The state includes: reinitializing the first cross-spectral low-pass filter based on the last two frames from the previous inactive encoding period status.

5. The method according to claim 3, wherein: The first cross-spectral low-pass filter is reinitialized based on coherence parameters from a previous inactive encoding period The state includes: reinitializing the first cross-spectral low-pass filter based on the second-to-last frame from the previous inactive encoding period status.

6. The method according to any one of claims 3 to 5, further comprising: The low pass filter is started (1901) during the DTX hold-off period. Updates.

7. The method according to any one of claims 1 to 6, wherein: Processing the audio input signal frame by frame includes processing the audio input signal frame by frame to generate a mono downmix signal, and encoding the motion content of each audio input signal includes encoding the motion content of the mono downmix signal.

8. The method according to claim 7, wherein: Processing the audio input signal frame by frame to generate the mono downmix signal comprises: processing the audio input signal frame by frame to generate the mono downmix signal and one or more stereo parameters, and encoding the activity content of the mono downmix signal comprises: encoding the activity content of the mono downmix signal and the one or more stereo parameters.

9. The method according to any one of claims 3 to 8, wherein: Determine according to the following formula: in, represents multiplication, is the low-pass coefficient, It is the frequency band The frequency coefficient set of is a vector containing the limits between the frequency bands, and is a complex number with absolute value = 1 and random phase.

10. The method according to claim 9, further comprising: Using the weighting function Weighted (2001).

11. The method according to claim 10, wherein: Using the weighting function The weighting is performed according to the following formula: in, is the discrete Fourier transform DFT energy spectrum of a mono signal which is a downmix of the audio input signal.

12. The method according to any one of claims 3 to 8, wherein: Determine according to the following formula: in, represents multiplication, is the low-pass coefficient, It is the frequency band The frequency coefficient set of is a vector containing the limits between the frequency bands.

13. The method according to claim 12, further comprising: Using the weighting function Weighted (2001).

14. The method according to claim 13, wherein: Using the weighting function The weighting is performed according to the following formula: in, is the discrete Fourier transform DFT energy spectrum of a mono signal which is a downmix of the audio input signal.

15. The method according to any one of claims 1 to 14, further comprising: The first frame having an inactive period of multiple frames does not update the , but updating the (2101).

16. The method according to any one of claims 1 to 14, further comprising: A dedicated cross-correlation estimation is performed (2201), the dedicated cross-correlation estimation being updated for the cross-spectrum only during the inactive period and / or during a DTX holdover frame, and the dedicated cross-correlation estimation is used for coherence estimation in the inactive period.

17. The method according to any one of claims 1 to 16, further comprising: The cross-spectral low-pass filter state is reset (2301) prior to any update in the DTX holdoff period or prior to any update in said inactive period.

18. The method according to any one of claims 1 to 17, further comprising: The low pass filter state is reinitialized (2401) at the beginning of a hold-off period or at the beginning of an inactivity period.

19. An encoder (400) adapted to enable the use of estimated coherence parameters to generate comfort noise in a network using discontinuous transmission (DTX), the encoder being adapted to: Receiving (1701) a time domain audio input comprising an audio input signal; The audio input signal is processed (1703) frame by frame by: encoding (1705) the active content of each audio input signal at a first bit rate until a period of inactivity is detected in the audio input signal; switching (1707) encoding from the active encoding content to the inactive encoding during the inactive period to encode the background noise at a second bit rate; estimating (1709) a coherence parameter during the inactive period based on low pass filtering of the cross spectrum or averaging of the cross spectrum, wherein estimating the coherence parameter comprises: initializing a low pass filter state of the cross spectrum based on coherence parameters from a previous inactive period; and The estimated coherence parameters are encoded (1711).

20. The encoder (400) according to claim 19, wherein The encoder is further adapted to perform a method according to any one of claims 2 to 18.

21. An encoder (400) adapted to enable the use of estimated coherence parameters to generate comfort noise in a network using discontinuous transmission (DTX), the encoder comprising: Processing circuit (1301); as well as A memory (1303) coupled to the processing circuit, wherein the memory includes instructions that, when executed by the processing circuit, cause the encoder to perform operations, the operations comprising: Receiving (1701) a time domain audio input comprising an audio input signal; The audio input signal is processed (1703) frame by frame by: encoding (1705) the active content of each audio input signal at a first bit rate until a period of inactivity is detected in the audio input signal; switching (1707) encoding from the active encoding content to the inactive encoding during the inactive period to encode the background noise at a second bit rate; estimating (1709) a coherence parameter during the inactive period based on low pass filtering of the cross spectrum or averaging of the cross spectrum, wherein estimating the coherence parameter comprises: initializing a low pass filter state of the cross spectrum based on coherence parameters from a previous inactive period; and The estimated coherence parameters are encoded (1711).

22. The encoder (400) according to claim 21, wherein The memory comprises further instructions which, when executed by the processing circuitry, cause the encoder to perform a method according to any one of claims 2 to 18.

23. A computer program comprising program code to be executed by a processing circuit (1301) of an encoder (400), whereby execution of the program code causes the encoder (400) to perform operations comprising: Receiving (1701) a time domain audio input comprising an audio input signal; The audio input signal is processed (1703) frame by frame by: encoding (1705) the active content of each audio input signal at a first bit rate until a period of inactivity is detected in the audio input signal; switching (1707) encoding from the active encoding content to the inactive encoding during the inactive period to encode the background noise at a second bit rate; estimating (1709) a coherence parameter during the inactive period based on low pass filtering of the cross spectrum or averaging of the cross spectrum, wherein estimating the coherence parameter comprises: initializing a low pass filter state of the cross spectrum based on coherence parameters from a previous inactive period; and The estimated coherence parameters are encoded (1711).

24. A computer program according to claim 23, comprising further program code to be executed by processing circuitry of the encoder, whereby execution of the further program code causes the encoder (400) to perform the operations of any one of claims 2 to 18.

25. A computer program product comprising a non-transitory storage medium, the non-transitory storage medium comprising program code to be executed by processing circuitry (1301) of an encoder (400), whereby execution of the program code causes the encoder (400) to perform operations comprising: Receiving (1701) a time domain audio input comprising an audio input signal; The audio input signal is processed (1703) frame by frame by: encoding (1705) the active content of each audio input signal at a first bit rate until a period of inactivity is detected in the audio input signal; switching (1707) encoding from the active encoding content to the inactive encoding during the inactive period to encode the background noise at a second bit rate; estimating (1709) a coherence parameter during the inactive period based on low pass filtering of the cross spectrum or averaging of the cross spectrum, wherein estimating the coherence parameter comprises: initializing a low pass filter state of the cross spectrum based on coherence parameters from a previous inactive period; and The estimated coherence parameters are encoded (1711).

26. The computer program product of claim 25, wherein: The non-transitory storage medium comprises further program code to be executed by processing circuitry (1301) of an encoder (400), whereby execution of the further program code causes the encoder (400) to perform operations according to any one of claims 2 to 18.

Images