Histogram equalization of contextual statistics of speech features for robust speech recognition.
Hsin-Ju HsiehBerlin ChenJeih-weih HungPublished in: Multim. Tools Appl. (2015)
Keyphrases
- speech recognition
- speech recognition systems
- speech signal
- speech synthesis
- noisy environments
- histogram equalization
- automatic speech recognition
- cepstral coefficients
- speech processing
- language model
- hidden markov models
- speech recognizer
- speaker identification
- mel frequency cepstral coefficients
- image enhancement
- isolated word
- pattern recognition
- speech recognition technology
- word error rate
- speech retrieval
- speaker independent
- speech recognition errors
- contrast enhancement
- speaker recognition
- gray level
- vocal tract
- image features
- feature space
- audio signal
- feature set
- co occurrence
- image processing