An intelligent platform to enhance customer service by letting robots to process customers’ requests
Naumen Erudite enables to create bots, which communicate with people using a natural spoken and typed language. You can use special interfaces to independently train and test bots as well as assess quality of their performance. The AI-based platform handles both voice calls and popular text channels, ensuring a unique omni-channel customer experience.
What tasks does Naumen Erudite solve?
NAUMEN intelligent systems deal with a wide range of activities: from classifying requests to providing full-scale customer services without human participation.
Classifying and intelligent routing of requests
Prompting dialogue scenarios to operators
Searching for branch office addresses, business units, ATM machines
Retrieving values from water and power meters
Ordering taxi cars
Paying orders
Opening bank accounts
Searching for evacuated cars
Tracking orders and parcels
Collecting debts
Consulting via a web site and a mobile application
Running NPS surveys
Scheduling delivery
Validating loan applications
Checking account balances
Enabling convenient human-to-bot communication
Why Naumen Erudite is optimal for high-quality servicing of customers without operators?
- Creates custom language models subject to specific features of your business.
- Is capable to simultaneously process different data types. The bots perceive both voice data (ASR) and typed data (DTMF)
- Deeply understands informal speech. Having received a request to call back, the bot asks for convenient time and correctly interprets replies like «Tomorrow after 3 o’clock», «By the end of this week», or «On Friday before lunch».
- Accurately identifies query topics and meanings. Asks clarifying questions If the bot is not sure what customers mean.
- Extracts facts from open remarks. For example, the bot extracts the following facts out of the phrase «Good day, I applied for an international passport the day before yesterday»: document — foreign passport; when — the day before yesterday.
- Produces no impact to customers’ experience. Customers can switch to operators at any time, if it is hard for them to speak with bots.
Naumen Erudite functional benefits
- Conducts free dialogues with customers in a natural language
- Uses modern technologies of machine learning and neural networks
- Adapts
ready-to-use language models to your terminology - Easily integrates with contact center automation platforms
- Supports
omni-channel mode
- Provides
user-friendly interface for training and setting up your bots - Gives a
drag-and-drop visual designer to develop servicing scenarios - Enables quick and situational additional training of bots
- Uses
pre-set algorithms to recognizeout-of-the topic phrases - Offers both working
on-premises and delivering through SaaS models
A unified platform for creating and controlling bots at all stages
Preparing training data and initial bot training
- An automatic aggregation of the dialogue history into clusters by key words
- Marking up training dialogues and assigning topics to messages in clusters
- Composing an hierarchical classifier of the service topics
Using a visual designer of service scenarios
- Creating service scenarios by topics of customers’ requests
- Applying basic bot skills of switching to the relevant scenario after extracting all significant facts from customers’ phrases thanks to usage of ML/DL algorithms
A pre-set recognition of human behavior patterns
Recognizing phrases beyond the main topic, which may occur at any stage of scenarios.
Specifying correct replies to the forms of:
- Greeting
- Goodbye
- Consent
- Rejection
- Displeasure
- 12 more replies beyond the context of the dialogs
Monitoring service efficiency of bots
- Online mode: using a customized dashboard to control metrics of robotiс services
- Offline mode: building detailed reports on running bots by sessions and service scenarios
Managing service quality of bots
- Correcting dialogues, in which bots committed errors
- Additional training of bots based on corrected data
Success Stories
The Russian Post
Voice bots classify and route customer requests, track the statuses of mailings, search for branch offices by street addresses or zip codes and give their operation hours.
After robotic automation request processing costs (package tracking) reduced by 6 times.
30 operators started to service more complex projects.
20% of calls are serviced without operator participation.
Mosenergosbyt
The bot accepts meter readings in voice mode, informs about account balance and files applications for paid services. It also calls around debtors, informs about debt collection reasons and agrees due dates with customers.
As a result of robotization:
Operator workload reduced by 40%
Debt payments increased by 15%
Promsvyazbank
A virtual assistant works with customers from small and medium size businesses on the bank’s website and through the mobile application. A bot replies to frequently asked questions, provides basic technical support and other services, such as: reserves accounts, selects service tariffs or merchant acquiring conditions.
The virtual assistant deals with 37 most frequent topics as of June 2019.
Inter RAO UES
The voice bot identifies request topics, customers by their addresses or by personal account numbers. It is capable of receiving customer’s readings of heat and water supply meters within a single call. The bot converts spoken data to numbers and transmits them to the billing system of OMSK RTS.
The robotic bot was put into service in 2 months from the project start.
Novosibirskenergosbyt
A voice bot accepts readings from power supply, hot and cold water meters, informs about last payments and account balance. The bot services not only individuals, but also legal entities, which call to consult on service conditions or check debts.
As of early October the bot serviced 150 000 requests
Moscow City Department of Information Systems
The voice bot of the Moscow Transport contact center helps Moscow residents to find an evacuated car and informs about the address of a special parking area. Upon receiving the customers’ requests, the bot explains where a car is and how to get it back depending on evacuation reasons.
The bot processes 80% of calls to the car evacuation hotline during working hours.
The bot services Moscow residents 2 times faster than the operators.
Achievements
Voice Bot Implementation Project based on Naumen Erudite reached the final stage of Crystal Headset 2019 contest
Voice Bot Implementation Project based on Naumen Erudite for the Russian Post customer service appears in the nomination “Best Technology Innovation: Self-Service Solution”
The Russian Post Robotiс Automation Project based on Naumen Erudite was acknowledged in the nomination “Best Use of Artificial Intelligence in Customer Service”
The Project “Naumen Erudite Voice Bots in the Russian Post Customer Support Center” recognized in the nomination “Artificial Intelligence”
Naumen Ecosystem
As a rule, companies implement bots as a combination of several different software solutions. They want bots to be able to work with text and voice, service as many different channels as possible.
As opposed to other suppliers of IT solutions, NAUMEN is able to deliver almost all functionality as a part of a unified solution that is based on its own software products. NAUMEN product line embraces all required components, such as:
- an
omni-channel switching and routing platform; built-in integration tools to connect multiple digital channels;- a full set of interfaces to service calls by human agents via IVR as well as by text and voice bots without human participation.
Contacts
Floor 5, bldg 3, 35B Vorontsovskaya St, Time Center Business Center,
Moscow 109147 (Krestyanskaya Zastava, Proletarskaya metro stations)
Phone/fax: