Intelligent Emotion Prediction from voice for call centers

Project Title

Intelligent Emotion Prediction from voice for call centers

Project Area of Specialization Artificial IntelligenceProject Summary

Intelligent Emotion Prediction from voice for call centers (IEPVCC) is an application which will be used in call center to predict the emotions of customers from his/her voice. Companies can predict the statistics of customers that are happy, sad, angry, exited from the company’s service. And it will also help the Call Center agent to predict the emotions of customer (during call) and deal it accordingly to increase the customer satisfaction.

 As Call Center job is one of the toughest jobs in the world, and the major challenge for agent is to deal the customers when they are not in front of you. All you can receive from customer’s side is voice. It is very difficult to predict the mood and emotions of the customers. The IEPVCC will be used in call centers to predict the emotions of the customer and help the agent to deal customers according to their mood: If call center agent know about the emotions of the customer so he/she can adapt the tune and behavior according to customers emotions and it will increase the customer satisfaction.

As customer’s satisfaction plays very important role in any business. Companies do different type of surveys and spent a lot of budget to get the statistics of customer satisfaction. This cost can be reduced by using IEPVCC. This system will process the previous call records and by applying some data mining techniques to generate the graph that will show the number of customers that are happy, sad, angry, exited etc. from our service.

Project Objectives

1. To take care of the customer’s service whether they are happy or not from the service. It will distinguish between two states: “Agitation” which includes anger and happiness and “Calm” that includes normal states and sadness.
2. 24/7, 365 days in a year, availability of the service.
3.  companies take surveys for the customer satisfaction that is a complex task, our system will give them a complete graph or statistics through which they can easily analyze.
4. System will Predict on the run time the emotions of the customer that makes it easier for the agent to deal accordingly.

Project Implementation Method

1. Develop API for sending or recieing the data form server.
2. Apply AI Techniques
   • Preprocess the data
   • Train the modal using convolutional neural  network (CNN)
   • Pridict Class (Emotion) 
3. Generate graphs using matplotlib and other supporting Libraries
4. Develop Website for giving Interface
5. Integrate Back end system with front end Website
6. Deploy System on Azure Server

Benefits of the Project

1. Reduce the cost and efforts of surveys for the customer satisfaction as this system will give them a complete graph or statistics.
2. It will help in improving the customers’ service in the short period of time.
3. It help the agent by pridicting  emotions of the customer that makes it easier for the agent to deal accordingly.

Technical Details of Final Deliverable

A website that will be user friendly and easy to use for end users. It will record the sound/voice and show the emotions of  that person on computer screen.

Final Deliverable of the Project Software SystemCore Industry OthersOther IndustriesCore Technology Artificial Intelligence(AI)Other TechnologiesSustainable Development Goals Decent Work and Economic GrowthRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	67400
Graphics Card (NVIDIA GTX 1060 4GB)	Equipment	1	34000	34000
BOYA microphone	Equipment	1	2800	2800
Azure membership cost (6 months)	Equipment	6	4600	27600
Thesis printing	Miscellaneous	3	1000	3000