Enhanced long-term predictor for Unified Speech and Audio Coding.
Jeongook SongHyen-O OhHong-Goo KangPublished in: ICASSP (2011)
Keyphrases
- long term
- audio stream
- audio visual
- linear predictive coding
- broadcast news
- speaker identification
- short term
- audio signals
- linear predictive
- emotion recognition
- text to speech
- cepstral features
- digital audio
- audio recordings
- speech recognition
- audio features
- linear prediction
- coding scheme
- speech processing
- multi stream
- neural network
- speech music discrimination
- multi modal
- audio video
- prosodic features
- acoustic signals
- multimedia
- automatic speech recognition
- spoken documents
- automatic transcription
- acoustic features
- voice activity detection
- visual information
- speaker diarization
- speech signal
- spontaneous speech
- audio files
- gaussian mixture model
- speech synthesis
- spoken language
- visual data
- signal processing
- speaker recognition