Please enable JavaScript.
Coggle requires JavaScript to display documents.
Paralinguistic characteristics of the English language …
Paralinguistic characteristics of the English language
Paralinguistics: Study of non-linguistic means of vocal and non-vocal communication, for example, acent, pitch, volume speech rate, modulation, tone of voice, fluency, intonation, body language, gestures, facial expressions, etc.
The paralinguistic properties of speech play an important role in human communication
Paralinguistic applies not only to speech but to writing and sign language as well
Paralinguistic Analysis - An Overview: Language was conceived as part of semantics whch deals with denotation
Segmentation: Speech is composed of components that can be segmented on the Time Axis from sub-phonics/phonemes to syllables/morphemes to words phrases and eventually narratives.
Phonation: Concerns the generation of acoustic energy
Larynx: Concerns by the action of vocal folds.
Laryngelistation: (irregular phonation, creak, vocal fry, creaky voice or pulse register) shows regular voiced stretches of speech
!
Metalanguage: People consistently associate different tones of voice with certain speaker personality characteristics
Sex and dialect: Display effects on phonetic characteristics such as central vowels, speech rate, flapping
Demographic traits: included gender, sex, age, education level, ethnicity, and geographic region
Speaker classification: Dialect, foreign accent, sociolect, age, gender, and medical conditions
Teaching
/ Learning
Therapeutic serious games: Allow games with a therapeutic purpose based on the verbal interaction and techniques of role-play
Tutoring systems: Employed to adapt to the system and the teaching pace. Essential for tutors and students to make learning successful
Speech and language resources
Well-defined test, develop, and training partitions without prototypical selection "friendly" cases of classification.
High diversity of speaker age, gender, ethnicity, language, etc.
Speech databases used for training and adaptation comprise the stored audio of speech for model learning (training) and testing
Speech databases
Audio speech for model learning, testing, and transcription of the spoken content
Acoustic chunking
Voiced/unvoiced parts, phonemes, syllables, words, sub-turns in the sense of syntactically and semantically motivated chunks below the turn level, or complete turns
Classification / regression
Enhanced recognition performance for the individual task, multi-task learning has recently attracted a lot of attention in the machine learning community
The first paralinguistic challenge: age, gender, and affect.
Single-words with the aim to learn more about the relative importance of acoustic-phonetic features for automatic speaker age recognition
Schotz used classification and regression three in two studies on Swedish.
