System and method for demographic analytics based on multimodal information

Abstract

The system and method of the present invention are described for automatic detection of error in the entry of particular category of individuals, (especially referring to gender and age classification) either real time while creating a database of such information or on an existing database on the record of individuals by analyzing their biometric characteristics (like speech, image or face) and other related demographic information (like name of the individual) in order to accord each individual with a unique identification.

Description

technical field [0001] The present invention relates to the field of data processing, and more particularly, to methods and devices for personal gender verification based on multimodal data analysis methods. Background technique [0002] In order to capture a record for an individual who conforms to the unique identifier of the individual, all necessary information needs to be collected and managed in an appropriate database. This information includes their name, gender, age, marital status, any photographs and biometrics (such as fingerprints, palm prints, retinal recognition, iris scans, facial recognition or voice samples). Such a valuable piece of information is stored in an appropriate database for further identification of individuals and verification of their gender. [0003] However, it has been observed that, in many cases, an individual's gender or age is incorrectly entered into such databases when record comparisons are made in real time. This in turn requires ...

AI Technical Summary

Problems solved by technology

However, identification based on current speech recognition typically exhibits high error rates; 98% accuracy is reported for clean speech and 95% accuracy for noisy speech

Furthermore, speech recognition systems work well under laboratory conditions, but tend to show considerable degradation in recognition rates when used in normal work environments

This loss in accuracy is likely to occur due to the unpredictable and variable noise levels found in normal working settings and the way individuals change their speech patterns to compensate for this noise

[0006] Incorporating name into one of the multiple parameters for gender and / or age identification and verification also poses multiple challenges based on personal geographic origin or location, and is thus prone to an attack error of approximately 5%

Images