A key feature of wireless mesh networks is that multiple independent paths through the network are available.Multiple descriptions coding is often suggested as a source coding scheme to take advantage of this path diversity.We compare multiple description (MD) coding with path diversity (PD) against a full-rate single description (SD) coder without PD, and two simple PD methods of 1) repeating a half-rate SD coder over both paths and 2) repeating the full-rate parent SD coder over the two paths. We first present a theoretical analysis comparing the average distortion per symbol in packetized communication using the above mentioned MD and PD methods to transmit a memoryless Gaussian source over additive white Gaussian noise channels. Next, using two new MD speech coders with balanced side descriptions derived from the AMR-WB and G.729 standards, we evaluate delivered voice quality using PESQ-MOS and compare MD coding against the PD methods for random and bursty packet losses. Both the theoretical analyses and the speech coding experiments show that with packet overheads, the simple PD methods may be preferable to MD coding. A new performance measure that incorporates both quality and bit-rate is shown to account for the tradeoffs more explicitly.

No abstract

No abstract

In this paper, we demonstrate the efficacy of transfer learning and continuous learning for various automatic speech recognition (ASR) tasks using end-to-end models trained with CTC loss. We start with a large pre-trained English ASR model and show that transfer learning can be effectively and easily performed on: (1) different English accents, (2) different languages (from English to German, Spanish, Russian, or from Mandarin to Cantonese) and (3) application-specific domains. Our extensive set of experiments demonstrate that in all three cases, transfer learning from a good base model has higher accuracy than a model trained from scratch. Our results indicate that, for fine-tuning, larger pre-trained models are better than small pre-trained models, even if the dataset for fine-tuning is small. We also show that transfer learning significantly speeds up convergence, which could result in significant cost savings when training with large datasets.

Abstract-We propose a new multiple description (MD) coder design based on the Adaptive Multi-Rate Wideband (AMR-WB) coder that can support transcoding-free communication between an ad-hoc network and another network that supports the AMR-WB codec. The encoder of the MD coder consists of the standard AMR-WB coder and a bit-stream splitting block that splits the AMR-WB bit-stream into two balanced descriptions. The decoder consists of a bit-stream substitution block that substitutes the missing bits, when only one description is received, to construct a valid AMR-WB frame that can be decoded using the standard AMR-WB decoder. We show that the performance of the new MD coder is better than a previous non-transcodingfree MD coder based on AMR-WB when transcoding is required and it is significantly better than using a single description of AMR-WB over the ad-hoc network supporting transcoding-free communication. I. VOICE OVER AD-HOC NETWORKSMobile Ad-hoc Networks (MANETs) are formed by mobile wireless hosts without the need of an existing infrastructure, unlike wireless cellular systems which require a centralized control and support system at the base station. Most of the wireless systems deployed today are centralized systems, wherein the nodes connected to the network communicate through an access point or a base station. MANETs do not need such a centralized support and can be deployed any place where nodes want to talk to each other or someone else through their neighbors. Since the nodes now depend on each other for communication, mobility of nodes implies that the routes for data transfer between two nodes are not fixed.MANETs are seen as future networks for Personal Area Networks (PAN), military environments, emergency operations and in an office or conference environments. Voice communication is essential in all of these scenarios. Real time communication over a MANET is a challenging problem because of the many transient characteristics of the network that arise due to the flexibilities that a MANET offers. The constantly changing routes between the sender and receiver nodes, lack of time synchronization between nodes in the network, broken links and delays involved in establishing Path diversity can be used to improve end-to-end connectivity between the nodes in a MANET. Using a path diversity scheme not only improves fault tolerance but also reduces overall packet losses and end-to-end delays. However, sending multiple copies of the same packet is inefficient usage of bandwidth. To improve the efficiency of using multiple paths, a source coding diversity method like Multiple Description (MD) coding can be used. In MD coding, multiple descriptions/bit-streams of the source are created in such a way that each description can be used to reconstruct the source with acceptable quality and two or more descriptions can be combined to give a better quality reconstruction.With different communication networks using different speech codecs, merging of the networks requires cross tandeming / transcoding of speech, w...