Exploiting Low-dimensional Structures to Enhance DNN Based Acoustic Modeling in Speech Recognition.
Pranay DigheGil LuyetAfsaneh AsaeiHervé BourlardPublished in: CoRR (2016)
Keyphrases
- speech recognition
- speech recognition systems
- low dimensional
- speaker independent
- speech recognizers
- hidden markov models
- language model
- acoustic models
- speech synthesis
- pattern recognition
- high dimensional
- automatic speech recognition
- noisy environments
- speaker identification
- noisy speech
- speech recognition technology
- speech processing
- dimensionality reduction
- speech recognizer
- speech understanding
- speech signal
- computer vision
- data points
- keyword spotting
- training process
- signal processing
- mobile devices
- speaker recognition
- speaker adaptation
- speech recognition errors
- n gram