We propose a simple, yet novel, multi-layer model for the problem of phonetic classification. Our model combines the frame level transformation of the acoustic signal with the segment level transformation via a temporal pooling architecture to compute class conditional probabilities of phones. Without the use of any phonetic knowledge, our model achieved the state-ofthe-art performance on the T...

This paper presents a generalized i-vector representation framework with phonetic tokenization and tandem features for text independent as well as text dependent speaker verification. In the conventional i-vector framework, the tokens for calculating the zeroorder and first-order Baum-Welch statistics are Gaussian Mixture Model (GMM) components trained from acoustic level MFCC features. Yet bes...

The aim of this study was to test whether both phonetic and phonological representations of intonation can be rapidly modified when imitating utterances belonging to a different regional variety of the same language. Our main hypothesis was that tonal alignment, just as other phonetic features of speech, would be rapidly modified by Italian speakers when imitating pitch accents of a different (...

The present paper investigates perceptions of speaker social-indexical information, including gender, age and social-class, from smaller phonetic segments such as gender-correlated phonetic variants. Since fundamental frequency (F0) is not the only cue to speaker gender identification, the perceptions are examined using gender-ambiguous sounding speech. The results of the study show that while ...

A phonetic representation of a language is used to describe the corresponding pronunciation and synthesize the acoustic model of any vocabulary. A phonetic representation with smaller phonetic units such as SAMPA-C for Mandarin Chinese and decision trees for parameter sharing are broadly applied to deal with the problem of large numbers of recognition units. However, the confusable phonetic rep...

Phonetic imitation, also called phonetic convergence, is currently at the heart of numerous investigations since it can inform us on both the nature of lexical representations and the link between production and perception processes in spoken language communication. A task that has been largely used to study phonetic imitation is the shadowing task, in which participants merely listen to and re...

A Conditional Random Field is a mathematical model for sequences that is similar in many ways to a Hidden Markov Model, but is discriminative rather than generative in nature. Here we explore the application of the CRF model to ASR processing by building a system that performs first-pass phonetic recogintion using discriminatively trained phonetic attributes. This system achieves an accuracy le...

This chapter describes the broad phonemic transcription in the CGN. First a broad overview of phonetic annotations in Dutch corpora is provided and a number of crucial dimensions are discussed: the source of annotation (human or automatic), the type of material involved, the level of transcription and the symbol set and transcription conventions. These dimensions serve as a guide through a numb...

Recognition and detection of non-lexical or paralinguistic cues from speech usually uses one general model per event (emotional state, level of interest). Commonly this model is trained independent of the phonetic structure. Given sufficient data, this approach seemingly works well enough. Yet, this paper addresses the question on which phonetic level there is the onset of emotions and level of...

One of the central findings of speech perception is that identical acoustic signals can be perceived as different speech sounds depending on adjacent speech context. Although these phonetic context effects are ubiquitous in speech perception, their neural mechanisms remain largely unknown. The present work presents a review of recent data suggesting that spectral content of speech mediates phon...