INTERSPEECH 2022 Audio Deep Packet Loss Concealment Challenge

ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

IEEE International Symposium on Signal Processing and Information Technology, 2013

This paper presents a packet loss concealment (PLC) method with applications to VoIP, wideband audio broadcast and streaming. The problem of modeling of time-varying frequency spectrum in the context of PLC is addressed and a novel solution is proposed for tracking and using the temporal motion of spectral flow. The proposed PLC utilizes a time-frequency motion (TFM) matrix representation of the audio signal where each frequency is tagged with a motion vector estimate. The spectral motion vectors are estimated by cross-correlating the movement of spectral energy within sub-bands across time frames. The missing packets are estimated in TFM domain and inverse transformed to the time-domain. In order to compare the proposed method with common approaches, objective performance evaluation of speech quality (PESQ), and subjective listening test in terms of MOS scores are conducted in a range of packet loss from 5% to 20%. The results demonstrate that the proposed algorithm improves performance compared to a number of benchmark methods including that of the lTV.

IEEE Sixth International Symposium on Multimedia Software Engineering, 2004

• Voice-over-IP: status and problems • Loss concealment problem -Previous work -IP voice traffic loss characteristics • Loss concealments for low bit-rate coded speech -Parameter-based MDC -Improving MDC quality • Summary • Future outlook

2006 IEEE 24th Convention of Electrical & Electronics Engineers in Israel, 2006

Packet loss in internet audio streaming applications is a well-known and difficult problem. Unless concealed, each such loss produces an annoying disturbance. In this work we present a new algorithm for audio packet loss concealment, designed for MPEG-Audio streaming, based only on the data available at the receiver. The proposed algorithm reconstructs the missing data in the DSTFT (Discrete Short-Time Fourier-Transform) domain using either the GAPES (Gapped-data Amplitude and Phase Estimation) or the MAPES (Missing-data Amplitude and Phase Estimation) algorithm. The algorithm was implemented on an MP3 coder but is also suitable for MPEG-2/4 AAC. Examined subjectively by a group of listeners, the proposed algorithm performs better than other previously reported methods, even at high loss rates.

2003

International Journal of Scientific and Engineering Research

The work consists of developing two MELP coders operating respectively at 1.2 kbps and 2.4 kbps. The improvements of the coder operating at 2.4 kbps by implementing packet loss concealment (masking) techniques based on the receiver which consists of interleaving information frames. These techniques are compared and mentioned on [1]. For this purpose, we extended this comparison to a method previously developed in our laboratory, called multiple descriptions coding (MDC) described on [2], which use two MELP coders operating respectively at 2.4 kbps for the first description and 1.2 kbps for the second description. We used as evaluation technique a method called PESQ (Perceptual Evaluation of Speech Quality) standardized by ITU-T.

Proc. International Conference on Spoken …, 2004

In this paper, we investigate the possibility of predicting lost packets for packet loss concealment using n-gram predictive models. Unlike the conventional repetition-based algorithms, the proposed algorithm is based on the Shannon game, which serves as a principle for predicting the speech parameters of lost packets using the previously received parameters. During training phase, we construct statistical backoff n-gram models. In test phase, the models are used to predict the speech parameters of lost packets. Experiments were performed on switchboard telephone speech database and the proposed algorithm is compared with the conventional repetitionbased algorithm. The performance is evaluated in terms of the spectral distortion between the original and the predicted (or repeated) speech. The algorithm based on the backoff n-gram models reduces the spectral distortion by 8.7% over the conventional repetition-based algorithm for the first lost packet after receiving one. Further it maintains about 6.2% improvement up to six consecutive lost packets. In terms of perplexity of the predictive models, backoff n-gram approach outperforms the repetition-based algorithm by 8.65%, which is very close to the improvement rate obtained from the spectral distortion measurement. 1

This paper describes new time domain techniques for concealing packet loss in the new 3GPP Enhanced Voice Services codec. Enhancements to the existing ACELP concealment include methods such as improved pitch prediction that may be guided as well as increased flexibility and accuracy of pulse resynchronization. Furthermore, the new method of separate LP filter synthesis aims for sound quality improvement in case of multiple packet loss, especially for noisy signals. An other enhancement consists of a guided LP concealment approach to limit the risk of creating artifacts during recovery. These enhancements are also used in the presented advanced TCX concealment method. Subjective listening tests show that quality is significantly increased with these methods.

2010

In this paper we review two Packet Loss Concealment (PLC) techniques, focusing on voice over packet networks. The first technique is based on Sinusoidal Modeling and the second one is based on Adaptive Lattice Modeling. Both techniques exploit look-ahead samples in order to improve their performance. In VoIP applications, look-ahead samples are commonly available from jitter buffers, without additional algorithmic delay over the jitter buffer inherent delay. The target location for these techniques is at the decoder side eliminating interoperability and standardization overhead. The proposed techniques were found to be more immune to high packet loss rate. Keywords-Packet Loss Concealment; Voice over Packet; Sinusoidal Modeling, Lattice Modeling

A major problem in low-latency Audio over IP transmission is the unpredictable impact of the underlying network, leading to jitter and packet loss. Typically, error concealment strategies are employed at the receiver to counteract audible artifacts produced by missing audio data resulting from the mentioned network characteristics. Known concealment methods tend to achieve only unsatisfactory audio quality or cause high computational costs. Hence, this study aims at finding a new low-cost concealment strategy using simplest algorithms. The proposed system basically consists of an period extraction and alignment module to synthesize conceal-ment signals from previous data. The audio quality is evaluated in form of automated measurements using PEAQ. Furthermore, the system's complexity is analyzed by drawing the computational costs of all required modules in all operating modes and comparing its computational load versus another concealment method based on auto-regressive modeling.