18187077. OUTPUT FROM ACOUSTIC INPUT simplified abstract (Ford Global Technologies, LLC)

OUTPUT FROM ACOUSTIC INPUT

Organization Name

Ford Global Technologies, LLC

Inventor(s)

Huanyi Shui of Ann Arbor MI (US)

Rajesh Gupta of Auburn Hills MI (US)

Gurram Sujith Kumar of Sullurpeta (IN)

Devesh Upadhyay of Canton MI (US)

Rajeev Kalamdani of Canton MI (US)

Douglas K. Grimes of Commerce Twp. MI (US)

Saumuy Puchala of Ypsilanti MI (US)

OUTPUT FROM ACOUSTIC INPUT - A simplified explanation of the abstract

This abstract first appeared for US patent application 18187077 titled 'OUTPUT FROM ACOUSTIC INPUT

Simplified Explanation:

The system described in the patent application involves a computer and memory that can transform acoustic data into an order spectrum, which is then inputted into a decoder to determine a feature vector. This feature vector can be classified as anomalous or non-anomalous by a one-class classifier, with the classified order spectrum being outputted.

• The system involves transforming acoustic data into an order spectrum.
• A decoder is used to determine a feature vector from the order spectrum.
• A one-class classifier is employed to classify the order spectrum as anomalous or non-anomalous.
• The classified order spectrum is then outputted.

Key Features and Innovation:

• Transformation of acoustic data into an order spectrum.
• Utilization of a decoder to determine a feature vector.
• Classification of the order spectrum as anomalous or non-anomalous using a one-class classifier.

Potential Applications:

The technology can be applied in various industries such as security, healthcare, and telecommunications for anomaly detection and classification.

Problems Solved:

The system addresses the need for efficient and accurate classification of acoustic data for anomaly detection.

Benefits:

• Improved accuracy in classifying acoustic data.
• Enhanced anomaly detection capabilities.
• Potential for increased security and efficiency in various industries.

Commercial Applications:

Potential commercial uses include security systems, healthcare monitoring devices, and telecommunications networks. The technology could have significant market implications in terms of improving anomaly detection and classification processes.

Questions about Acoustic Data Classification:

1. How does the system transform acoustic data into an order spectrum? 2. What are the potential real-world applications of classifying acoustic data as anomalous or non-anomalous?

Original Abstract Submitted

A system is disclosed that includes a computer and memory, the memory including instructions to transform acoustic data to an order spectrum and input the order spectrum to a decoder to determine a feature vector. The feature vector can be input to a one-class classifier to classify the order spectrum as anomalous or non-anomalous and the classified order spectrum can be output.