Google boasts several artificial intelligence announcements that it intends to make public soon during its Cloud Next conference, which is set to start in San Francisco. Most of them are based on the entity’s democratization of machine learning resources technologies.
In fact, during its keynote announcement, Google revealed that its AutoML Vision tool would be currently available in public beta after culminating an alpha duration, which kick-started back in January. The start of the alpha period was marked by the unveiling of the Google’s Cloud AutoML initiative.
Cloud AutoML serves as a technique that allows non-experts, particularly those without coding fluency or machine learning expertise, to train their self-learning models using technologies that are available as part of Google’s cloud computing provision. The first tool was AutoML Vision, which allows you to build a machine learning model for both object and image recognition.
With AutoML Vision’s entry into public beta, it will be available to many organizations, researchers, and enterprises, especially those that find such an AI useful but lack the expertise or resources to create their own training models.
More often, companies could easily use AI software via an applicable API such as the Cloud Vision API that Google offers to third parties. Nonetheless, Google is creating its Cloud AutoML tools to serve companies, mainly those that fall outside the technology sector and have specific requirements that need training on custom data.
Google also announced two new Cloud AutoML domains whereby one is meant for natural language while the other for translation. The company’s ability to break down both spoken and written words using software creates the foundation of its Google Assistant product.
Also, the competency of its translation algorithms, which are AI-trained, has allowed Google Translate to be so successful across various types of languages.
Although it is difficult to create complex software and models such as those belonging to Google without the right sizable datasets, resources, and expertise, Google is simplifying the beginning of basic training of custom models thanks to the new domains.
The company said that leading publishing company Hearst is utilizing AutoML Natural Language to assist in tagging and organizing content across its various magazines and many international and national versions of such publications.
Additionally, Google granted AutoML Translation to Nikkei Group, a Japanese publisher, which is known for publishing and translating articles across different languages daily.
Aside from its new Cloud AutoML domains, Google is also creating an AI-driven customer service agent that serves as the first human-sounding voice that a caller first comes into contact with over the phone.
The product dubbed Contact Center AI will be used together with Google’s Dialogflow package, which offers tools to enterprises for creating conversational agents.
Thanks to Contact Center AI, Google appears to be moving into an area where callers are more familiar with the perception of interacting with a robot and are doing so out of their own will through interacting with customer service proactively.
Looking at this context, it seems that there is a high likelihood that the technology will dominate the operations of call centers in the future.