In this week’s roundup, learn how artificial intelligence (AI) technology can detect COVID-19 by the sound of a cough. Understand how AI can empower sales and marketing teams, and discover how innovators are using machine learning (ML) to track the pandemic’s effect on mental health, democratize data in the Library of Congress, and help artists create audiovisual content.
By Aylin Woodward, contributing writer for BusinessInsider.com
Researchers at MIT have developed an AI-powered model that can detect whether someone is infected with coronavirus simply by the sound of their coughs. The technology can identify small changes in the sound of coughs that the human ear can’t hear. The AI had incredibly high accuracy, picking up the infections in 98.5% of the people with coronavirus infections—even if they were asymptomatic and forced their coughs. The researchers hope to turn it into a mobile app to aid in real-time testing.
by David A. Teich, contributing writer for Forbes.com
AI is helping sales and marketing teams close the loop on identifying, qualifying, and closing leads. While CRMs have become much more robust over the past decade, AI models have the power to analyze every customer and lead touchpoint and predict buyer behavior. Experts predict this technology will significantly improve sales and marketing ROI.
By Anne Trafton, contributing writer for News.MIT.edu
In an effort to explore the mental health impacts of the pandemic, MIT researchers used natural language processing (NLP) to analyze Reddit posts. The technology looked for the frequency of words related to anxiety, depression, and other topics that indicate mental health issues. After using the machine-learning technique to analyze 800,000 Reddit posts, the researchers found that the use of troubling language doubled from January to the height of the first wave of the pandemic in April. The research is intended to identify the most vulnerable groups of people and help ensure they have access to mental health resources.
By Stepen Gosset, contributing writer for BuiltIn.com
The Library of Congress is the largest library in the world, and its collections of print material have been largely digitized, resulting in massive datasets. However, because the data is largely siloed into separate archives, researchers have found it difficult to analyze the data at scale. So, the Library recently launched an initiative to implement machine learning to ensure researchers can access and analyze the library’s data across collections at scale.
Traditionally, creating audiovisual content has been time-consuming and expensive. A new Chilean startup is looking to change that by using AI and machine learning. The platform, dubbed Runway, has the ability to create and edit audio and visual content using AI-powered algorithms. While still in the startup stage, the company has already worked with New Balance in the design of a shoe and Grammy-nominated band YACHT in creating music for their latest album.
Did you see an interesting article in the last week? Share it with us! Send it to astuttle [at] lityx.com.