Search

This research proposes an enhanced measurement method for VoIP quality assessment which provides an improvement to accuracy and reliability. To improve the objective measurement tool called the simplified E-model for the selected codec, G.729, it has been enhanced by utilizing a model of Mean Opinion Score (MOS), called a subjective MOS prediction model, based on native Thai users who use the Thai-tonal language. Then, the different results from the simplified E-model and subjective MOS prediction model were used to create the Bias function, before adding to the simplified E-model. Finally, it has been found that the outputs from the enhanced simplified E-model for the G.729 codec show better accuracy when compared to the original simplified E-model, specially, after the enhanced model has been evaluated with 4 test sets. The major contribution of this enhancement is that errors are reduced by 58.87% when compared to the generic simplified E-model. That means the enhanced simplified E-model as proposed in this study can provide improvement beyond the original simplified one significantly. [ABSTRACT FROM AUTHOR]

Usage of Automatic Speech Recognition (ASR) systems is increasing day-by-day for voice centric applications in mobile handheld and Voice over Internet Protocol (VoIP) devices. The necessity is also increasing to find out the ASR performance under different network impediments. Among them, speech and audio coding standards is the one, which affects the ASR performance greatly, when, using them with different sampling and bit rates in the practical systems. Another common impediment which influences the ASR accuracy is the bit errors in the wireless networks and packet drop conditions in the VoIP networks. ASR performance with some of the speech coding standards under noise conditions for the wireless networks is reported in the literature. However, each study is reporting the ASR performance for few narrowband codecs with different speech databases and different ASR toolkits like RAPHEL, HTK, SPHINX, etc. In this paper, the analysis on ASR performance while using both narrowband and wideband speech and audio coding standards, which are currently accepted for GSM mobile and VoIP networks, using the common speech database-TIMIT, and using ASR toolkit-SPHINX, is presented. The Mean Opinion Score (MOS), which is the generally accepted speech quality measurement technique, is also analyzed for all the speech and audio coding standards, using the same speech database. The results of the studies carried out for the ASR word accuracies and MOS values for different narrowband and wideband speech and audio codecs under no-loss conditions are presented. Results for different rates of packet drop condition which is the common noise scenario in wired networks such as VoIP (which is also merging with wireless networks) are also presented. The observation is that though some of the codecs are showing poor MOS performance at lower bit rates, the corresponding ASR performance is comparable with other codecs at higher bit rates. [ABSTRACT FROM AUTHOR]

A common communications convergence scenario which is being adopted in personal communications relates to the combination of wireless and cellular networks by the use of multimode terminals. Since most of the wireless networks were initially dimensioned only for data communications, this paper shows how voice over wireless LAN dimensioning could be addressed under the optimal network throughput and the perspective of voice quality, using a simple approach. The maximum number of simultaneous users resulting from throughput is limited by the collisions taking place in the shared medium with the statistical contention protocol. The voice quality is conditioned by the delay and the packet loss in the contention protocol. Both approaches are analyzed within the scope of the voice codecs commonly used in voice over wireless LANs, to conclude that voice dimensioning based on network throughput and voice quality show complementary results. Additionally the use of low rate codecs in voice over wireless LANs is advantageous for the network performance point of view but may produce poor voice quality results. Mid range codecs like G729 could represent a trade-off for quality throughput. For these reasons, voice quality and wireless network throughput have to be taken into account in the network admission control, design and deployment to ensure a satisfactory user experience. The impact of handoff interval of wireless convergent networks on the conversation quality need also be assessed for a proper network design. [ABSTRACT FROM AUTHOR]