4. AUDIO ANALYTICS ON EDGE Flashcards
15 Questions
Which of the following is not a valid time domain feature?
a. Root Mean Square Energy
b. Zero Crossing Rate
c. Spectral Centroid
d. Amplitude Envelope
Spectral Centroid
What is the typical audio sampling rate?
a. 44.1 kHz
b. 20 kHz
c. 88 kHz
d. 22 kHz
44.1 kHz
Which of the following is an important feature for speech recognition?
a. All of the above
b. Edges
c. Histogram of gradients
d. Spectral features
Spectral features
Which type of input will be best suited as an input to a machine learning algorithm?
a. Raw audio signal
b. Spectrogram of audio signal
c. Spectrum of audio signal
d. Filtered audio signal
Spectrogram of audio signal
What is the importance of Mel-scale?
a. Sounds of equal distance on the Mel Scale are perceived to be of equal distance to humans
b. It is easier for a machine learning algorithm to tackle
c. It is convenient to use
d. It is a form of compression of audio signals
Sounds of equal distance on the Mel Scale are perceived to be of equal distance to humans
Which of the following sound frequencies will a normal human able to distinguish better?
a. 10000 to 10200 Hz
b. 1000 to 1200 Hz
c. 5000 to 5200 Hz
d. 100 to 200 Hz
100 to 200 Hz
What is the need of a power spectrogram?
a. To capture the change of power of a signal with time
b. None of the above
c. To capture the change of frequency of a signal with time
d. To capture the change of amplitude of a signal with time
To capture the change of frequency of a signal with time
What is the typical fundamental frequency range of adult voice?
a. 31 to 19000 Hz
b. Below 85 Hz
c. 85 to 155 Hz
d. 165 to 255 Hz
31 to 19000 Hz
What is Cepstrum?
a. It is a tool to find anomalies in the signal
b. All of the above
c. It is a tool to find frequencies in the signal
d. It is a tool for finding spectrum of a spectrum
It is a tool for finding spectrum of a spectrum
Why YAMNet is a suitable sound classification model for edge applications?
a. It is a decision tree based approach
b. It uses multiple sound features as input
c. None of the above
d. It is based on light weight MobileNet v1
It is based on light weight MobileNet v1
Speech signals can be considered as which of the following type?
a. Stationary signals
b. Non-stationary signals
c. Random signals
d. None of the above
Non-stationary signals
Why audio analytics on the edge is an important task?
a. Sensitivity of the data is critical for many applications
b. Redundancy of the data can be reduced (enable compression)
c. Low latency is critical for fast action (e.g. audio scene classification)
d. All of the above
All of the above
What does a spectrum of a signal provide?
a. A measure of noise in the signal
b. All of the above
c. A measure of frequency components present in the signal
d. A measure of amplitude of the signal
All of the above
Which filter type attenuates frequencies higher than the cutoff point?
a. Low-Pass
b. Notch
c. High-Pass
d. Band-Pass
Low-Pass
Why are Mel-frequency cepstral coefficients (MFCC) commonly employed in speech processing?
a. All of the above
b. It emphasize features of the audio signal that are important for human speech
c. It reduces the redundancy in speech processing
d. It is one of the most efficient representations of speech for employing as an input to machine learning
All of the above
