نتایج جستجو برای: speaking accuracy

تعداد نتایج: 367019  

2001
Hiroaki Nanjo Kazuomi Kato Tatsuya Kawahara

The paper addresses large vocabulary spontaneous speech recognition focusing on acoustic modeling that considers the speaking rate. Using the real lecture speech corpus collected under the priority research project in Japan, we have made baseline acoustic model, and evaluated on the automatic transcription of oral presentations by experienced speakers and obtained word accuracy of 58.2%. Compar...

2002
Kozo Okuda Tatsuya Kawahara Satoshi Nakamura

In this paper, we propose a speaking rate compensation method using frame period and frame length adaptation. Our method decodes an input utterance using several sets of frame period and frame length parameters for speech analysis. Then, this method selects the best set with the highest score which consists of the acoustic likelihood normalized by frame period, language likelihood and insertion...

Journal: :Journal of speech, language, and hearing research : JSLHR 2005
Puisan Wong Richard G Schwartz James J Jenkins

The present study investigated 3-year-old children's perception and production of Mandarin lexical tones in monosyllabic words. Thirteen 3-year-old, Mandarin-speaking children participated in the study. Tone perception was examined by a picture-pointing task, and tone production was investigated by picture naming. To compare children's productions with the adult forms, 4 mothers of the children...

1989
Patti Price

SRI is developing a multi-modal interface to an air travel database that will permit cooperative planning via interactive human-machine problem solving. The project goals are real-time, large vocabulary (3000 words), high semantic accuracy (90%), speaker-independent understanding of spontaneously spoken natural language in a restricted domain. The grammar should be shown to be habitable and the...

2004
Hiroshi Fujimura Katsunobu Itou Kazuya Takeda Fumitada Itakura

The dependency of conversational utterances on themode of dialogue is analyzed. A speech corpus of 800 speakers collected under three different modes, i.e., talking to a human operator, an WOZ system and an ASR system, is used for analysis. Some characteristics such as sentence complexity loudness of the voice and speaking-rate are found to be significantly different among the dialogue modes. L...

2015
Matthew W. Wagers Emily Pendleton

Broadly speaking, subject relative clauses are easier to understand than object relative clauses (Wanner & Maratsos, 1978, Ford, 1983, Holmes & O'Regan, 1981, among many others). This can be demonstrated by a variety of methods that measure the speed and accuracy with which sentences containing relative clauses are processed. As an example, compare (1a), which contains a subject relative clause...

2002
Michael Israel

Modern uses of the word literally are surveyed, and the solecistic misuse of this word as an expression of speaker commitment is shown to be a natural semantic extension of its basic metalinguistic meaning. The development appears to be typical of a large class of intensifiers, though it is argued that literally itself is only in the early stages of such an evolution. Aspects of this developmen...

Journal: :Current Biology 2003
Nigel Williams

Many have pondered the potential impact of Mendel's work if Darwin had come across it in his lifetime. Darwin found difficulty with a blending inheritance and natural selection but a new letter suggests he was grappling with a more modern view. Nigel Williams reports.

This study intended to develop a model describing speaking strategies for EFL learners by taking into account the effects of learners’ gender and proficiency on the application of strategies. Accordingly, this study was planned to have two main analyses, namely qualitative and quantitative. In this respect, 30 EFL learners' viewpoints were sought, and then, based on the elicited responses, a 21...

نمودار تعداد نتایج جستجو در هر سال

با کلیک روی نمودار نتایج را به سال انتشار فیلتر کنید