Abstract
Biometric technology is spearheading the existing authentication methods in the IoT. Considering the balance between security and convenience, voice biometrics seems to be the most logical biometric technologies to be used. The authors present an extensive survey to identify, analyze, and compare various methods and algorithms for the different phases in the process of speaker identification/recognition, which is the part and parcel in voice biometrics. The chapter is intended to provide essential background information to those interested in learning or planning to design voice authentication systems. The chapter highlights the need for a biometric authentication system, the reason why we prefer voice, its present state of affairs, and its scope with fog computing to be used in IoT.
TopBackground
Evolution
Recognition of voice started when introducing the first automatic voice recognition device in 1952.This non computerized machine was capable to recognize human voice pronouncing single digits. Later in 1992 the first prototype of a modern voice recognition system, Sphinx-II, was created which could perform real-time voice recognition and became suitable for using in modern software application.
In 1867, for instance, Alexander Melville Bell laid the groundwork for future voice biometrics research by inventing a language called Universal Alphabetic. Using this system, which replicates the position a mouth makes when speaking a certain speech pattern, it’s possible to transcribe not only what a person is saying, but how they are saying it (“A Brief History of Voice Biometrics - VoiceVault Voice Authentication,” n.d.)
The first use cases of voice biometric identity verification emerged by the introduction of machines called spectrographs during World War II. It was used by the American soldiers to intercept voice transmissions and track enemy movements. It didn’t provide accurate results as it was primitive at the time. Later the first modern voice biometrics engine capable of accurately registering and determining an end user’s voiceprint was created by Texas Instruments in 1976.
Key Terms in this Chapter
Liveness Detection: The detection of whether an acquired biometric sample data is coming from a currently active live user.
Internet of Biometric Things: An architecture that supports the accessibility of applications in IoT through biometric authentication schemes.
Speaker Recognition: The process of recognizing a speaker from his own voice characteristics.
Speaker Diarization: The process of identifying and labeling different speakers from a speech segment by segmentation and clustering methods.
Voice Morphing: The process of generating a person’s voice using software in order to impersonate or obscure his identity.
Behaviometrics: Derived from the two words behavior and biometrics , behaviometrics means the analysis of a person’s behavior rather than physical characteristics in order to identify him uniquely.
Multimodal Biometric Authentication: Refers to the use of more than one physiological or behavioral biometric characteristic for authentication purposes.