Voice Degradation in Using Speech Recognisers for Transcribing Inventory Data.
Final technical rept. Sep 85-Nov 86,
LEEDS UNIV (ENGLAND) INST FOR TRANSPORTATION STUDIES
Pagination or Media Count:
Problems arise in long sessions of voice tape-recording for off-line data entry to computers via speech recognition systems, as a result of operator fatigue or loss of attention. In this study the task of reading vehicle license plates aloud for one hour was simulated in laboratory conditions, each speaker undergoing on recording session with feedback conditions, each speaker undergoing one recording session with feedback on recognition accuracy and one without. No significant difference in recognition success rates between the two conditions was analysed acoustically for fatigue-induced changes both in long-term prosodic characteristics including fundamental frequency, intensity, spectral balance and rate and in segmental characteristics such as frequency of occurrence of different sound types and segmental durations. Although no consistent tendencies were found in all speakers. It is concluded although, the choice of speakers and conditions resulted in the more sophisticated techniques developed during the project on the basis on the automatic segmentation of continuous speech is capable of more revealing analysis than the relatively crude techniques used previously.
- Voice Communications