Endpoint Detection of Isolated Korean Utterances for Bimodal Speech Recognition in Acoustic Noisy Environments.
Hyun-Hwa OhHong-Seok KwonJong-Mok SonKeun-Sung BaeSung-Il ChienPublished in: ISMIS (2003)
Keyphrases
- speech recognition
- noisy environments
- endpoint detection
- speech recognition systems
- speaker independent
- speech recognizers
- noisy speech
- acoustic models
- speaker verification
- acoustic features
- speech recognizer
- speech signal
- automatic speech recognition
- speaker identification
- background noise
- speech enhancement
- hidden markov models
- digit recognition
- language model
- visual speech
- pattern recognition
- speech synthesis
- natural language
- speech retrieval
- word recognition
- audio signal
- speaker recognition
- sound source
- wavelet transform
- neural network
- feature extraction