Utterance clustering is one of the actively researched topics in audio signal processing and machine learning. This study aims to improve performance utterance by multichannel (stereo) signals. Processed signals were generated combining left- right-channel a few different ways then extracting embedded features (also called d-vectors) from those processed applied Gaussian mixture model for super...

This paper proposes a combination of source-normalized weighted linear discriminant analysis (SN-WLDA) and short utterance variance (SUV) PLDA modelling to improve the short utterance PLDA speaker verification. As short-length utterance i-vectors vary with the speaker, session variations and phonetic content of the utterance (utterance variation), a combined approach of SN-WLDA projection and S...

The paper presents a series of experiments on speech utterance classification performed on the ATIS corpus. We compare the performance of n-gram classifiers with that of Naive Bayes and maximum entropy classifiers. The n-gram classifiers have the advantage that one can use a single pass system (concurrent speech recognition and classification) whereas for Naive Bayes or maximum entropy classifi...

This paper presents the results of an analysis of user reactions towards system failures in turn-taking in human-computer dialogues. When a system utterance and a user utterance start with a small time difference, the user may stop his/her utterance. In addition, when the user utterance ends soon after the overlap starts, the possibility of the utterance being discontinued is high. Based on thi...

It is unclear how the prosodic features in long spontaneous speech change. We focused on the F0 declination of an utterance to investigate the F0 transition by taking into consideration the length and syntactic structure of a Japanese utterance. In the case of an utterance without a strong syntactic boundary, the F0 range is fixed to a certain width, and the F0s decline toward the end of the ut...

A notion of truth as applicable to events of fact-stating use (utterances) of a sentence type is arguably presupposed and required by our evaluative practices of the use of language. The truth of an utterance clearly depends on what the utterance says. This fundamental dependence seems to be captured by the schema that if an utterance u says that P , then u is true iff P . Such a schema may thu...

PURPOSE Evidence from children's spontaneous speech suggests that utterance length and utterance position may help explain why children omit grammatical morphemes in some contexts but not others. This study investigated whether increased utterance length (hence, increased grammatical complexity) adversely affects children's third person singular -s production in more controlled experimental con...

A relation between utterance speed and the amount articulatory behavior for each phoneme reduces is investigated on the basis of the articulatory data taken by using a magnetic sensory system. The relation between the utterance speed and the change of the utterance timing for each phoneme is also focused on. Two linear-regression models are proposed: one predicts reduced articulatory movements ...