Tkool Electronics

Finally, it is important to know how the echo canceller behaves in the double-talk situations—when both talkers talk simultaneously. Obviously, the parties prefer to hear each other throughout the entire conversation and hear little or no echo during double-talks. Therefore, the echo canceller's double-talk performance should also be addressed when designing and testing echo cancellers or simply choosing which one to integrate to the phone.

resistor color codes

Finally, it is important to know how the echo canceller behaves in the double-talk situations—when both talkers talk simultaneously. Obviously, the parties prefer to hear each other throughout the entire conversation and hear little or no echo during double-talks. Therefore, the echo canceller's double-talk performance should also be addressed when designing and testing echo cancellers or simply choosing which one to integrate to the phone.

Knowledge built in the back room or from unstructured information sources is often outdated before it even reaches your agents and customers, so it is crucial to constantly capture the experience of your front-line agents during the support process. Take the age-old example of buying a new car: as soon as you drive it off the lot, it loses its value. As soon as your knowledge base goes live, it's bound to contain some outdated information, so it's important to capture and reorganize the knowledge based on every new customer interaction or experience.

During a typical customer service call, 70% of the time is spent diagnosing the problem and 30% of the time is spent in delivering the answer. By optimizing troubleshooting and diagnostic questions and reordering them based on their past usage, call handle time metrics can improve dramatically.

resistor color codes

Experience-Based

It is almost impossible, due to repetition, for agents not to learn the answers to the most common questions that are asked each and every day. Agents learn from experience. But the average search engine can only give, at best, some weight to the most common problem-solution sets. In order to be effective in a customer service environment, a search engine must be able to utilize the experiences and knowledge of its users in order to provide accurate results. A feedback mechanism must be available in order for the system to know if a particular solution solved a customer's issue.

Intuitive/Self-Learning Search

resistor color codes

Although it is great to deliver search results based on experience, what if you've never seen a particular issue before? You wouldn't want a system that returns zero results or just says, Sorry can't find an answer!”. It's important for the technology to leverage past experiences to provide a best guess” as to where to begin troubleshooting the problem. This most closely mimics the way your subject matter experts (SME) handle calls.

If you can't resolve a problem and you ask a SME, they will typically ask you a few more qualifying questions, then they will either say — Ah, I've seen that before — this is what you do to fix it” or they will say I've never seen that happen in this particular context (i.e., same platform, application, version, etc.), so let's try this solution that worked in a similar situation.” If it does in fact work, the SME will make a mental note (experience) that it works in that environment also. Search engines that can provide this type of intuition are self-learning — they learn from existing experiences and apply those experiences to similar problems.

resistor color codes

The Bottom Line

Simple search engines might be adequate for the average Web surfer, but not for customer service agents. In a support environment, where accurate answers must be delivered quickly to satisfy your customers, search technologies embedded within agent-facing knowledge management tools must be more than fast and easy-to-use. They must provide a systematic process that enables agents to pinpoint the most relevant solutions.

Implementing TDM over CSMA/CA required the development of advanced algorithms that control every access point in the network to eliminate contention, collisions and interference — even while operating on the same channels. But the results speak for themselves, Kohli said.

Meru's Air Traffic Control architecture brings predictability through scheduling — on top of content management — to reduce jitter and latencies, to the extent that we can support up to 100 users on any given access point,” he said. Up to 30 of those can be voice calls.”

That's five times the user density and voice-handling capability of the base 802.11b network, though proprietary schemes do somewhat better, Kohli added. However, those proprietary schemes require software additions to the client,” he said. Ours doesn't.”

The company's Over-the-Air QoS scheme offers fine-grained QoS per application, user and flow in both the up- and downlink between the client and the access points. While some plans, such as one announced recently by Airflow Networks, split the voice and data between channels, Meru's approach allows a single channel to carry both.

Meru's system comprises a $595 access point based on a radio and baseband from RF Micro Devices Inc., the company's own MAC and an $8,000 central controller that interfaces to the enterprise backbone and the network's already installed switches, such as Cisco Systems' Catalyst line. The controller also implements what Kohli calls the basic requirements of centralized security enforcement and ready deployment and management.

访客,请您发表评论:

Powered By Tkool Electronics

Copyright Your WebSite.sitemap