E-Book, Englisch, 344 Seiten, E-Book
Schuller / Batliner Computational Paralinguistics
1. Auflage 2013
ISBN: 978-1-118-70662-6
Verlag: John Wiley & Sons
Format: EPUB
Kopierschutz: Adobe DRM (»Systemvoraussetzungen)
Emotion, Affect and Personality in Speech and Language Processing
E-Book, Englisch, 344 Seiten, E-Book
ISBN: 978-1-118-70662-6
Verlag: John Wiley & Sons
Format: EPUB
Kopierschutz: Adobe DRM (»Systemvoraussetzungen)
This book presents the methods, tools and techniques that arecurrently being used to recognise (automatically) the affect,emotion, personality and everything else beyond linguistics('paralinguistics') expressed by or embedded in humanspeech and language.
It is the first book to provide such a systematic survey ofparalinguistics in speech and language processing. The technologydescribed has evolved mainly from automatic speech and speakerrecognition and processing, but also takes into account recentdevelopments within speech signal processing, machine intelligenceand data mining.
Moreover, the book offers a hands-on approach by integratingactual data sets, software, and open-source utilities which willmake the book invaluable as a teaching tool and similarly usefulfor those professionals already in the field.
Key features:
* Provides an integrated presentation of basic research (inphonetics/linguistics and humanities) with state-of-the-artengineering approaches for speech signal processing and machineintelligence.
* Explains the history and state of the art of all of thesub-fields which contribute to the topic of computationalparalinguistics.
* C overs the signal processing and machine learning aspects ofthe actual computational modelling of emotion and personality andexplains the detection process from corpus collection to featureextraction and from model testing to system integration.
* Details aspects of real-world system integration includingdistribution, weakly supervised learning and confidencemeasures.
* Outlines machine learning approaches including static, dynamicand context-sensitive algorithms for classification andregression.
* Includes a tutorial on freely available toolkits, such as theopen-source 'openEAR' toolkit for emotion and affectrecognition co-developed by one of the authors, and a listing ofstandard databases and feature sets used in the field to allow forimmediate experimentation enabling the reader to build an emotiondetection model on an existing corpus.
