An embedded stereo speech and audio coding method based on principal component analysis.
Mao-shen JiaChangchun BaoXin LiuXiao-ming LiRu-wei LiPublished in: ISSPIT (2011)
Keyphrases
- coding method
- principal component analysis
- audio stream
- audio visual
- coding scheme
- speaker identification
- mel frequency cepstral coefficients
- broadcast news
- bit rate
- audio features
- compression ratio
- speech recognition
- dimensionality reduction
- speech signal
- shape coding
- multimedia
- computer vision
- feature extraction
- gaussian mixture model
- low dimensional
- face recognition
- stereo images
- face images
- lossless coding
- three dimensional
- sparse coding
- depth map
- inter frame
- higher order
- feature vectors
- acoustic features
- image segmentation
- image quality
- image processing