Speech compression with preservation of speaker identity.
John LeisMark PhythianSridha SridharanPublished in: ICASSP (1997)
Keyphrases
- speech recognition
- speaker recognition
- automatic speech recognition
- audio visual
- speaker verification
- speaker identification
- prosodic features
- speaker dependent
- automatic speech recognition systems
- speaker diarization
- speech signal
- vocal tract
- compression algorithm
- image compression
- speech synthesis
- speech sounds
- hidden markov models
- speech recognizer
- vector quantization
- synthesized speech
- audio stream
- text to speech
- automatic transcription
- phoneme recognition
- compression scheme
- speaker adaptation
- data compression
- spoken language
- spontaneous speech
- compression ratio
- broadcast news
- speaker independent
- language model
- identity management
- feature extraction
- multi modal
- visual information
- visual data
- compression rate
- gaussian mixture model
- acoustic features
- acoustic models
- image quality
- probabilistic neural network
- speech recognition systems
- emotion recognition
- dialogue system
- endpoint detection
- non stationary
- mel frequency cepstral coefficients
- hearing impaired
- visual speech