| 摘要 |
A method and apparatus for identifying any one of a plurality of words, each word having an audible form represented by a sequence of spoken speech elements picked up by microphone (16). This system includes a CPU (22) for controlling system operation, a first unit (24) for storing the data representing the plurality of words, a second unit (26) for storing the data used to calculate probabilities. The spoken speech elements are converted into an analog signal and converted into digital signals by interface (10), and interpreter (30) is connected to interface (10) to receive and interpret the digital signals to produce an identification of each spoken speech element, and a calculating unit (32) uses the probabilities produced from data in unit (26) and compares the interpreted speech elements with the data stored in unit (24) in order to identify the word represented by the stored data which corresponds to the spoken speech elements.
|
