Evaluating the robustness of privacy-sensitive audio features for speech detection in personal audio log scenarios.
Sree Hari Krishnan ParthasarathiMathew Magimai-DossHervé BourlardDaniel Gatica-PerezPublished in: ICASSP (2010)
Keyphrases
- audio features
- audio visual
- low level
- feature set
- audio stream
- visual features
- privacy sensitive
- speaker identification
- genre classification
- acoustic features
- music genre classification
- music information retrieval
- multi modal
- text data
- sound source
- audio signal
- audio content
- privacy preserving
- real world
- high level
- image processing
- medical data
- mel frequency cepstral coefficients
- visual information
- face recognition
- feature selection
- training data