Vector quantizationwith renormalized splits for wideband speech.
Miguel Arjona RamírezPublished in: DSP (2011)
Keyphrases
- speech signal
- speech recognition
- speech quality
- speaker identification
- feature vectors
- automatic speech recognition
- speaker recognition
- vocal tract
- speech synthesis
- language acquisition
- signal to noise ratio
- multimedia
- sparse matrix
- automatic speech recognition systems
- recognition engine
- endpoint detection
- data sets
- spoken language
- linear prediction
- non stationary
- hidden markov models
- vector space
- human computer interaction
- multi modal
- multiscale
- text to speech synthesis
- hearing impaired
- information retrieval