| Popis: |
There have been dramatic improvements during the last decade in the performance of both automatic speech recognition (ASR) systems and the computers on which they run. Speaker independent recognition with an active vocabulary of over 1,000 items is now possible even under the difficult conditions imposed by the public switched telephone network. The impressive raw recognition power of commercial recognizers has led to the development of a wide range of complex information, call management, and transaction services that rely on ASR. Despite this progress, it has become clear that recognition performance in the lab is not a good predictor of success in the field, and that extensive work on dialogue design and human factors "tuning" is required before most services can be used successfully by the general population. Since recognizer capabilities strongly influence user-interface design, and since these capabilities have been changing rapidly over the past decade, there is no clear body of knowledge that a designer can turn to when developing new services that rely on speech. |
