An Ultra-low Power RNN Classifier for Always-On Voice Wake-Up Detection Robust to Real-World Scenarios.
Emmanuel HardyFranck BadetsPublished in: CoRR (2021)
Keyphrases
- real world
- text localization and recognition
- wide range
- partial occlusion
- feature selection
- recurrent neural networks
- synthetic data
- landmark detection
- nearest neighbor
- support vector machine classifier
- false negative
- automatic detection
- false positives
- neural network
- driver assistance systems
- spam detection
- detection accuracy
- classification algorithm
- detection method
- object detection
- case study
- detection rate
- detection algorithm
- voice activity detection
- training data
- support vector machine
- class labels
- support vector
- feature space
- linear classifiers
- data mining
- multiple classifiers
- vehicle detection
- knn
- learning classifier systems
- false alarms
- low power
- anomaly detection
- face detection