The voice recognition accuracy of the AI voice bot "Leo Kani" for building names has doubled!

By utilizing address data, the voice recognition accuracy of the AI voice bot "Leo Kani" for building names has doubled!

Could you tell us about the overview of the service?

A voicebot is an automated voice response system that utilizes AI technology. It analyzes incoming voices using technologies such as speech recognition and natural language processing, and then provides answers using speech synthesis technology. Voicebots primarily handle customer service tasks such as product orders and reservation requests in call centers, replacing human operators.
Since it can also collect customer information such as name and address, the AI voice bot "Leo Kani" can handle all phone inquiries.

"Leo Kani" was developed after five years of research, focusing on high speech recognition accuracy and flexibility, utilizing millions of data points including names, company names, and addresses for learning and meticulous tuning. We are currently continuing to add to the database and are striving to further improve recognition accuracy from its current 97.4% rate.

For user companies that adopt "Leo Kani" it will help address needs and concerns such as "We want to be able to handle work outside of business hours, such as at night or in the early morning" and "We need extra staff due to a sudden increase in demand for TV shopping or large-scale advertising."

*AI voice bot Leo Kani 1.0 provides voice bot services based on Nuance Japan's speech recognition solution, Nuance voice platform, the cloud service Genesys Cloud CX, and Amazon Web Services.

Please tell me about the functions of our product, and its advantages and effects.

For voice recognition systems used in call centers, the key is how to improve the recognition rate of customer addresses, building names, and other similar information.
Previously, we were using our own original speech recognition engine, but we had issues with address recognition data. After introducing address and building name data from GeoTechnologies, Inc., the speech recognition rate for building names doubled.

Please tell me about your implementation track record.

We have a proven track record of implementations in pharmaceutical companies, manufacturers, mail-order companies, and IT companies.

Could you tell us about the circumstances and deciding factors that led you to adopt our products?

Previously, we used our own original speech recognition engine, but due to issues with address recognition data, we introduced address and building name data from GeoTechnologies, Inc..
The deciding factors were the implementation cost, the amount of data, and the speed of the sales response.

Could you tell us about your future plans?

We are conducting research and development activities for next-generation voice bots that utilize generative AI.

 

Click here for the website of Kanden CS Forum Co., Ltd.

Click here for the 'Leo Kani' homepage.

For inquiries about MapFan DB, please click here.