Consumers have indicated that knowledge, friendly support, and first-call resolution are the primary (dis)satisfiers with your brand when they contact you.
Turn your contact center into an action-driven instrument using Xdroids OperatorAssist; the real-time Voice- and EmotionAnalytics solution. Capture, analyze and recommending actions are all done in just a fraction of a second. Agents are advised on meeting their compliance, track customer emotion, get pop-ups with relevant information, automatically schedule a follow-up call if needed and can wrap-up with an automatically generated call transcript.
Frequently asked questions from Contact Center Managers.
How can I provide the best possible support to agents while they are in a conversation so that they can give the best experience to customers?
My agents need help on 100% of their calls, how can my supervisors immediately interact and help them?
How can I upgrade customer satisfaction and maximize sales efficiency at the same time?
Can we reduce the agent’s wrap-up time by providing a call-summary at the end of the call?
OperatorAssist can help your agents with quality and compliance checks and to give meaningful advice during a conversation with customers. As a result, the confidence-level of the agent, as well as the overall call quality, will increase.
Xdroid’s AI algorithms support the engine so it can advise agents with selecting the most appropriate sales scripts or keywords. We use the information we get from emotions, keywords, data from past conversations, data from 3rd party systems (e.g., CRM) to provide personalized assistance in each call.
For more details, a list of functionalities and visualization examples, please scroll down.
Make your agents so much more efficient by eliminating wrap-up time.
See how Xdroid's real-time VoiceAnalytics can capture a summary of every call that can be copied straight into CRM and other benefits in this video our UK Distributor, avoira produced.
When we designed our next generation engine, we wanted to make sure it transformed the capabilities real-time analytics can offer to contact center agents. One of our customers said:
“The system is only limited by one’s imagination.”
Our next generation of machine learning and artificial intelligence engine have implemented the most advanced deep neural network technologies. We didn’t only want to be focused on very fast speech processing: we took the end-to-end approach to provide optimum performance, the presentation layer is equally important as it is the interface to the agent
Our new engine can be used either in real-time or in post-processing mode. In both modes, it can handle Keyword Spotting (KWS) as well as Speech-to-text (STT). In post-processing mode it feeds our VoiceAnalytics solution, in the real-time mode we have two applications, real-time agent assistance; OperatorAssist, and a wrap-up screen.
Real-time agent assistance
During a call, our system informs agents of their speech style, compliance, length of silence periods in the conversation and other quality-oriented KPI’s. Keyword Spotting can be used to compare against specific criteria; this is called the ComplianceTracker. To help the agent say the right things, the ComplianceTracker coaches the agents with a script. This real-time coaching ensures that best practices can be shared throughout a companies agent population, providing higher success, customer satisfaction, and compliance rates.
Web support is triggered automatically based on the mention of specific keywords. When a keyword is detected, a link to supporting documents appears. In the accompanying picture, an example is given. Messages can be turned into alerts to inform the Contact Center Quality Manager about ongoing calls that might require immediate intervention based on speech characteristics or used keywords.
When a call is completed, the wrap-up screen allows for a summarize to save in the CRM. Elements of the wrap-up screen can be the overall emotion index, keywords or sentences, a compliance summary and any other element that could be used to follow the customer on its journey. The wrap-up screen could even contain a full speech to text transcript.
KWS vs. STT
Xdroid does both so that we can assess the benefits of each of the two methods. With full speech-to-text, a transcript of a conversation can be shared and analyzed across the organization. The question is, however, who is going to read those transcripts. A transcript requires less storage than a full audio conversation, which seems to be an advantage. Very few people would want to go through reading all sorts of transcripts unless it is a legal case (in that case you may want to be able to listen to the conversation and make absolutely sure what has been said). For most other purposes, for instance, analyzing which topics gave the most negative emotions during a call, you may prefer to do a keyword search combined with our TrueEmotion detection. With all these capabilities in one engine, you can select the technology that best suits the situation.
CPU vs. GPU
Xdroid uses GPU cards for data processing, and there is a good reason for doing so. In essence, SpeechAnalytics compares pictures of waves with each other and tries to find the best mapping of those pictures to determine which letter has been pronounced and it does this for each letter. For that, a graphical user card is much better equipped than a CPU card. Instead of trying to explain this elaborately, watch the Mythbusters video, and you will get the picture.