| 摘要 |
<p>The invention concerns a speech recognition device (1) comprising an audio processor (2) for acquiring an audio signal and a linguistic decoder (6) for determining a sequence of words corresponding to the audio signal. The linguistic decoder of the inventive device comprises a language model (8) determined from a first set of at least a syntactic block defined only by a grammar and a second set of at least a second syntactic block defined by one of the following elements, or a combination of said elements: a grammar, a list of sentence segments, a n-gram system.</p> |
