Mixtures of Deep Neural Experts for Automated Speech Scoring.
Sara PapiEdmondo TrentinRoberto GretterMarco MatassoniDaniele FalavignaPublished in: CoRR (2021)
Keyphrases
- network architecture
- speech recognition
- semi automated
- fully automated
- speech synthesis
- neural model
- neural network
- human experts
- text to speech
- speech signal
- mixture model
- domain experts
- recognition engine
- endpoint detection
- neural fuzzy
- speaker identification
- automated analysis
- scoring model
- blind source separation
- bio inspired
- audio visual
- recurrent neural networks
- multi modal
- signal processing
- principal component analysis
- natural language