This set of Cognitive Radio Multiple Choice Questions & Answers (MCQs) focuses on “Next Generation Wireless Network – User Cognitive Techniques – 1”.
1. Which among the following is an application of speech recognition?
a) Stress characterization
b) Network support
c) Access control
Explanation: Speech recognition supports access control by providing user authentication. Voice comparison can be used to establish the identity of a person. It can also be employed to identify communicating parties and personalization of devices.
2. Which among the following is the process of providing voice samples to generate a template in speech recognition?
Explanation: Voice samples are obtained from users to generate a model or a template. The verification phase involves the comparison of a signal taken from the application with that of a template for matching. The result of verification may be positive or negative identification.
3. Which among the following parameters influence errors that occur in speaker recognition?
a) Amplitude dependency
b) Pitch dependency
c) Text dependency
d) Rate of speech dependency
Explanation: Speaker recognition consists of two errors namely false recognition and recognition failure (failure to identify enrolled user). The systems are characterized by whether the speech is dependent on text involving phrase prompts or independent of text such as a conversational speech.
4. Which among the following categories include a badge for authentication?
Explanation: The major categories of user authentication are knowledge, traits, token, and behaviour. PINs and passwords are associated with knowledge. Keys and badges are associated with tokens. Behaviour involves usage patterns. Features such as voice and fingerprint are used to sense behaviour. An implementation including four categories offers strong protection.
5. Which among the following is false about behaviour based authentication?
a) Behaviour authentication supports continuous mode operation
b) Behaviour authentication includes recognizing patterns
c) Behaviour authentication includes biometric detection
d) Behaviour includes network interaction
Explanation: Behaviour based user authentication involves recognizing actions, interests, tendencies, and patterns. Behaviour based user authentication exhibits minimal intrusion to user activity, and is transparent. It supports continuous mode operation and requires a limited amount of resources.
6. Which among the following is not a disadvantage of adapting specialized hardware for processing?
a) Update management
b) Functionality modification
c) High bandwidth requirement
Explanation: Biometric based user authentication obtains input information through an input device such as a microphone, camera, etc. The information is processed by using software, or specialized hardware. Specialized hardware increases the complexity of introducing updates and modifying functionalities on the fly. Software processing offers lower security than specialized hardware.
7. Which among the following is a traditional method for language identification?
a) Phrase translation
b) Word translation
c) Letter translation
d) Phonetic translation
Explanation: Phonetic transcription is a traditional method for language identification. Phones are the subunits of basic sounds of speech. The relation between the occurrences of phones in different language is analyzed to develop a statistical model for each language for identification.
8. How to obtain cepstrum coefficients?
a) Cepstrum coefficients are obtained by taking Fourier transform of the spectrum
b) Cepstrum coefficients are obtained by taking Fourier transform of decibel spectrum
c) Cepstrum coefficients are obtained by taking Laplace transform of decibel spectrum
d) Cepstrum coefficients are obtained by taking Laplace transform of the spectrum
Explanation: Cepstrum coefficients are features employed in speech signal processing. It is obtained from the spectrum of a speech signal. Cepstrum coefficients are obtained by taking the Laplace transform of the decibel spectrum. Shifted delta cepstrum coefficients measure the changes in the speech spectrum over multiple frames of spectrum to model language characteristics.
9. Which among the following is not a region of interest associated with text to speech conversion?
a) Transcript generation
b) Pronunciation of named entities
c) Quality of voice synthesis
d) Conveyance of expression
Explanation: Text to speech conversion voices textual information. The source of the information could be text based communication or display readouts. The regions of interest are associating news with particular voice and tone, producing human sounding voice, pronunciation of named entities, quality of voice synthesis, and conveyance of expression.
10. Speaker stress characterization enables cognitive radio to assess the success of a decision on the basis of verbal cues.
Explanation: Speaker characterization is the process of finding the emotional state of a user through speech processing. Knowledge of stress state helps to improve situational awareness, identify irritation, determine stress level, and assess the success of a decision on the basis of verbal cues.
Sanfoundry Global Education & Learning Series – Cognitive Radio.
To practice all areas of Cognitive Radio, here is complete set of 1000+ Multiple Choice Questions and Answers.