Exploiting Jump-Resonance Hysteresis in Silicon Auditory Front-Ends for Extracting Speaker Discriminative Formant Trajectories.
Kenji AonoRavi Krishna ShagaShantanu ChakrabarttyPublished in: IEEE Trans. Biomed. Circuits Syst. (2013)
Keyphrases
- speech recognition
- automatic speech recognition
- vocal tract
- visual information
- signal processing
- discriminative training
- speaker verification
- high speed
- feature extraction
- low cost
- speaker recognition
- audio visual
- speech signal
- brain computer interface
- moving objects
- discriminative features
- trajectory data
- semi supervised
- speaker adaptation
- feature selection
- pattern recognition
- cmos technology
- speaker identification
- maximum likelihood
- phase transition
- steady state
- dimensionality reduction
- unsupervised learning
- markov chain