SensoSmart Virtual Cloud Sensors: What performance can we expect from VoIP?

Monday, March 17, 2014

What performance can we expect from VoIP?

http://www.voipmechanic.com/mos-mean-opinion-score.htm

'Codecs- how they affect MOS- Mean Opinion Score

A non-compressed codec, G711, will give the best voice quality, as no compression and decompression is required. This in turn makes this G711 less susceptible to packet loss, which even in small amounts will quickly degrade voice quality. Other codecs which use compression techniques consume less bandwidth, a good thing, but this compression process in itself lowers the voice clarity and introduces a delay. Additionally, compressed codecs are even more susceptible to packet loss. In addition G711 will maintain better quality with more latent connections, whereas compressed codecs' voice quality can quickly degrade as latency increase, as the chart below reflects.

Comparison of compressed and ono-compressed VoIP codecs and MOS values.

Comparison of voice quality for G711 & G729 codecs as latency increases.'

No comments:

Post a Comment

SensoSmart Story

Easy to use affordable personal devices including smart phones, notebooks, readers, and tablets have exploded in popularity. These devices connect people around the world with family, friends, and resources, for education, in the workplace or business, to receive medical care, for entertainment and many other services; yet there are still some people who are not able to fully access these resources. Modern devices can provide sophisticated sensors and applications with the potential to transform service and device experiences and extend reach to even more people. Our cognitive sensor project SensoSmart leverages existing infrastructure with new features to personalize the services, simplify control of assistant devices, improve human-robot-interaction, and to enable people to perform tasks beyond their own abilities.

We share some examples:

Many people are hard of hearing and benefit from hearing aids; approximately 4 out of 5 people who would benefit from hearing aids do not have them. Hearing aids are quite expensive (between 2 and 6 thousand dollars US) and are not typically covered by insurance. People with hearing loss can miss out on so many important things and are disconnected from enjoyable popular experiences of their peers. (Approximately 38 million Americans have significant hearing loss and many would benefit from hearing aids. Thirty to forty percent of people over 65 and fourteen percent of those between 45 and 64 have some type of hearing loss.)

We propose to introduce an affordable software hearing aid app downloadable to smart phones or accessible on the cloud to assist people who are hard of hearing so they can enjoy audio applications. This approach can increase the reach for audience and provide an enjoyable entertainment experience for many who would otherwise have no access to audio applications. Over time, we would like to extend the features to improve the experience of people who are hard of hearing when listening in a variety of audio situations. We will enhance our services with user profiles including audiograms, preferences, and genetics to personalize hearing algorithms and applications; we will conduct network analytics of the behavior of the users to measure app effectiveness and predict new needs of the users.

Another example:

Assistant robots and devices are expensive, limited in function, and can be complex to learn and use. SensoSmart virtual sensors provide a means to exchange data between devices so they can connect to shared resources learning across the cloud. SensoSmart provides a convenient means to share useful control programs and features among users of assistive devices through cloud and crowd sourcing. This enables lower cost models to benefit people with disabling conditions. SensoSmart provides an easy way to widely share robot functionality letting robots learn from one another and utilize popular strategies available on the cloud.

Stevens Institute of Technology Cognitive Sensing for Health