Improving speech processing trough social signals: automatic speaker segmentation of political debates using role based turn-taking patterns.
Fabio ValenteAlessandro VinciarelliPublished in: SSPW@MM (2010)
Keyphrases
- speech processing
- turn taking
- signal processing
- speech recognition
- speaker identification
- social interaction
- social media
- image segmentation
- noisy environments
- multi party
- multimedia systems
- artificial intelligence
- speech signal
- computer mediated
- automatic speech recognition
- gaussian mixture model
- e learning
- variable length
- computer vision
- language model
- hidden markov models
- pattern recognition
- image processing
- machine learning
- non stationary
- probabilistic model
- collaborative activities
- language identification
- social networks