This paper describes an efficient search algorithm for a high speed and high accuracy LVCSR system. A conventionally used lexical tree search is an efficient method, but has a problem in incorporating the language probability. To solve this problem, we propose in this paper a new efficient search algorithm incorporating the language model structure. In our developed LVCSR, 2-pass search algorit...

For a variety of theoretical and practical reasons, dictation is a promising general procedure for promoting both conscious learning and subconscious acquisition of L2 grammatical structures. Although traditional dictation tends to be rather boring, the method allows a number of interesting and motivating variant techniques. This article describes four such variants, useful and effective dictat...

This paper reports on the development and the initial evaluation of a dictation&spelling prototype exercise for second language (L2) learners of Swedish based on text-to-speech (TTS) technology. Implemented on an already existing Intelligent Computer-Assisted Language Learning (ICALL) platform, the exercise has not only served as a test case for TTS in L2 environment, but has also shown a poten...

We report a new type of dysgraphia, which we term dyscravia. The main error type in dyscravia is substitution of the target letter with a letter that differs only with respect to the voicing feature, such as writing "coat" for "goat", and "vagd" for "fact". Two Hebrew-speaking individuals with acquired dyscravia are reported, TG, a man aged 31, and BG, a woman aged 66. Both had surface dysgraph...

The article describes AE, a Hebrew-speaking individual with acquired dysgraphia, who makes mainly letter position errors in writing. His dysgraphia resulted from impairment in the graphemic buffer, but unlike previously studied patients, most of his errors related to the position of letters rather than to letter identity: 80% of his errors were letter position errors in writing, and only 7% of ...

The text input for non-alphabetic languages, such as Chinese, has been a decades-long problem. Chinese Dictation using large vocabulary speech recognition provides a convenient mode of text entry. In contrast to a character based Dictation system [5], a word-based Mandarin dictation system has been designed [3] (based on Apple's PlainTalk speech recognition technology [4]) for efficient entry o...

The paper reports the results of two studies on the spelling performance of 1st graders in a transparent writing system. The spelling performance of Italian children was assessed to determine the cross-task relationship between spelling to dictation and spontaneous spelling at the single word level (Study 1) and at the text level (Study 2), respectively. In study 1, 132 Italian children's spell...

We present a universal approach to uncover and correct systematic local errors in complex speech-to-text systems. Whereas previous work to minimize speech recognition errors mostly relies on N-best lists or word lattices, our approach is merely based on the first-best system output. The paradigm of Transformation-Based Learning (TBL) is adapted from tagging-like applications to themore complica...