Using Autoregressive Models for Real-Time Packet Loss Concealment in Networked Music Performance Applications

Sacchetto, Matteo; Huang, Yuen; Bianco, Andrea; Rottondi, Cristina

doi:10.1145/3561212.3561226

Cited by 8 publications

(9 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Connectivity interruptions may happen, so long as their frequency of occurrence is low enough for low-complexity error correction schemes to compensate. Such techniques include packet loss concealment methods [45]- [48]. Satisfying these KPIs is necessary to maintain a stable tempo and to ensure a satisfactory auditory perception, thus enabling synchronicity among performers and, more generally, a high-quality interaction experience [3,Ch.…”

Section: Systemsmentioning

confidence: 99%

Latency and Reliability Analysis of a 5G-Enabled Internet of Musical Things System

Turchet

Casari

2024

IEEE Internet Things J.

View full text Add to dashboard Cite

The availability of high-performance embedded audio systems, along with high-bandwidth and low-latency connectivity options provided by 5G networks, is enabling the Internet of Musical Things (IoMusT) paradigm. A central component of this paradigm is represented by Networked Music Performances (NMPs), where geographically displaced musicians play together over the network in real time. However, to date, IoMusT deployments over 5G networks remain scarce, and very limited statistical results are available on the actual latency and reliability of 5G networks for IoMusT and NMP scenarios. In this paper, we present a private 5G IoMusT deployment and analyze its performance when supporting NMPs. Our IoMusT system is composed of up to four nodes and includes different background traffic conditions. We focused on the assessment of the sole wireless link, as the measurements can be easily transferred to a realistic NMP architecture involving a WAN by compounding them with those of the WAN. Our results show that latency increases with the number of nodes and with the presence of background traffic, whereas the reliability did not vary with the complexity of the conditions. For all tested scenarios, the average measured latency was below 24 ms (including a jitter buffer of 10.66 ms), whereas packet losses occurred with a probability of less than 0.01. However, irregular spikes were found for all latency and reliability metrics, which can significantly reduce the quality of service perceived by the users of NMP applications. Finally, packet loss and latency resulted to be uncorrelated, which suggests that they have different root causes.

show abstract

Section: Systemsmentioning

confidence: 99%

Latency and Reliability Analysis of a 5G-Enabled Internet of Musical Things System

Turchet

Casari

2024

IEEE Internet Things J.

View full text Add to dashboard Cite

show abstract

“…We define the objective in ( 5) as the linear combination of time-domain mean squared error (7), spectral convergence (8), and the L 1 -norm of the regularized log-magnitude error (9). Similarly to prior work [12], [13], [28], [29], the spectrotemporal losses are evaluated at multiresolution scales, i.e., using Q = 3 different sets of Fourier analysis parameters as shown in Table 1.…”

Section: B Neural Network Trainingmentioning

confidence: 99%

“…2) comprises a skip connection passing through a convolutional layer with no dilation and two stacks of convolutions with dilation factor of two and four, respectively, batch normalization, and LeakyReLU with slope α = 0.2. The number of filters progressively grows in the encoder (8,16,32,64), and decreases symmetrically in the decoder (64, 32,16,8). The convolutional layers have filters of size 11 in the encoder and seven in the decoder.…”

Section: Neural Network Architecturementioning

confidence: 99%

“…G.711 [4]. Since audio signals are typically "well-behaved" within short-time windows, linear autoregressive (AR) models proved effective and relatively inexpensive for speech [5], [6], [7] and networked music applications [8]. More recent PLC methods involve the use of deep neural networks to predict future packets from the previously received audio context.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Hybrid Packet Loss Concealment for Real-Time Networked Music Applications

Mezza,

Amerena,

Bernardini

et al. 2024

IEEE Open J. Signal Process.

View full text Add to dashboard Cite

This work was partially supported by the European Union under the Italian National Recovery and Resilience Plan (NRRP) of NextGenerationEU, partnership on "Telecommunications of the Future" (PE00000001 -program "RESTART.") This work involved human subjects or animals in its research. The author(s) confirm(s) that all human/animal subject research procedures and protocols are exempt from review board approval.

show abstract

“…The latter may utilize innovative multisensory interfaces, both to enrich their own listening experience and to participate actively in the music creation process [3]- [5]. An increasing body of literature dealing with IoT-based musical devices [6]- [10], communication architectures and protocols [3], [11]- [16], packet loss recovery methods [17], [18], synchronization mechanisms [19], [20], studies involving distributed musicians [21], [22], as well as discussion papers [23], [24] confirms the growing interest of the IoT community about the Internet of Musical Things.…”

Section: Introduction a General Contextmentioning

confidence: 99%

Is Music in the Air? Evaluating 4G and 5G Support for the Internet of Musical Things

Vignati,

Nardini,

Centenaro

et al. 2024

IEEE Access

View full text Add to dashboard Cite

The full potential of the Internet of Musical Things (IoMusT) paradigm can be fully unleashed only in the presence of widespread, reliable wireless connectivity, allowing musicians to connect their smart instruments (almost) anywhere they are. For this reason, we propose a realistic, end-to-end communication architecture for a IoMusT system based on public fifth-generation (5G) mobile networks that considers a networked music performance use case, and we introduce a model for the resulting system. We define high-level service requirements and key performance indicators for the network's connect-compute architecture. We evaluate our solution via system-level simulations using the well-known 5G-LENA/ns3 and Simu5G/OMNeT++ frameworks. We found that operating 6 or more IoMusT devices over a fourthgeneration (4G) network results in a worst-case latency well over 20 ms for more than 90% of the packet transmissions, while losing more than 10% of packets. On the other hand, operating the same number of devices over a 5G network reduces the latency significantly. After testing our findings both in single-cell and in multi-cell scenarios, assuming a transient in the upgrade of mobile network infrastructures from 4G to 5G, we consider a E-UTRA-NR dual connectivity (EN-DC) scenario, where a 5G base station serves a subset of the users of a 4G cell. Furthermore, we conducted a user study where musicians were asked to assess their playing experience during simulated 4G-and 5G-based networked music performances. The results of the simulations and of the user study consistently indicate that the use of 5G technology improves performance significantly, and advocate the need for a 5G framework to fully support the IoMusT.

show abstract

Using Autoregressive Models for Real-Time Packet Loss Concealment in Networked Music Performance Applications

Cited by 8 publications

References 6 publications

Latency and Reliability Analysis of a 5G-Enabled Internet of Musical Things System

Latency and Reliability Analysis of a 5G-Enabled Internet of Musical Things System

Hybrid Packet Loss Concealment for Real-Time Networked Music Applications

Is Music in the Air? Evaluating 4G and 5G Support for the Internet of Musical Things

Contact Info

Product

Resources

About