Using bidirectional lstm recurrent neural networks to learn high-level abstractions of sequential features for automated scoring of non-native spontaneous speech.
Zhou YuVikram RamanarayananDavid Suendermann-OeftXinhao WangKlaus ZechnerLei ChenJidong TaoAliaksei IvanouYao QianPublished in: ASRU (2015)
Keyphrases
- recurrent neural networks
- high level abstractions
- long short term memory
- neural network
- spontaneous speech
- feed forward
- low level
- echo state networks
- recurrent networks
- feature vectors
- artificial neural networks
- image features
- co occurrence
- feature space
- feature extraction
- non stationary
- expert systems
- structural features
- multimedia
- artificial intelligence
- genetic algorithm