Employing Bottleneck and Convolutional Features for Speech-Based Physical Load Detection on Limited Data Amounts.
Olga EgorowTarik MrechNorman WeißkirchenAndreas WendemuthPublished in: INTERSPEECH (2019)
Keyphrases
- data sets
- data collection
- low level
- raw data
- data analysis
- high quality
- synthetic data
- end users
- knowledge discovery
- data quality
- speech recognition
- data processing
- image data
- data points
- data structure
- database
- data distribution
- data mining
- data mining techniques
- feature space
- feature selection
- sensor data
- machine learning
- complex structures