Deep bottleneck features and sound-dependent i-vectors for simultaneous recognition of speech and environmental sounds.
Sakriani SaktiSeiji KawanishiGraham NeubigKoichiro YoshinoSatoshi NakamuraPublished in: SLT (2016)
Keyphrases
- environmental sounds
- feature extraction
- acoustic features
- feature vectors
- sound source
- speech signal
- automatic speech recognition systems
- speech recognition
- pattern recognition
- speech recognition systems
- feature space
- classification accuracy
- feature set
- audio features
- object recognition
- dynamic environments
- visual features
- neural network
- image features
- control system
- video sequences
- face recognition