A speaker-independent word recognition based on HMM using orthogonalized phonetic segment codebook.
Hiroshi MatsuuraTsuneo NittaPublished in: Systems and Computers in Japan (1994)
Keyphrases
- speaker independent
- word recognition
- speech recognition
- automatic speech recognition
- hidden markov models
- noisy environments
- speaker identification
- handwriting recognition
- handwritten word recognition
- speaker dependent
- speech signal
- vector quantization
- digit recognition
- speech recognizer
- speaker adaptation
- character recognition
- cursive script
- language model
- pattern recognition
- acoustic models
- broadcast news
- word segmentation
- handwritten words
- speech recognition systems
- bag of words
- word level
- document analysis
- noise reduction
- non stationary
- feature vectors
- face recognition