Loading...
| Friend's email | |
| Your name | |
| Your email | |
| enter code | |
This page was sent successfuly
- Type of Document: M.Sc. Thesis
- Language: Farsi
- Document No: 40733 (19)
- University: Sharif University of Technology
- Department: Computer Engineering
- Advisor(s): Sameti, Hossein
- Abstract:
- In contrast to automatic speech recognition (ASR), which converts a speaker’s spoken utterance into a text string, spoken language understanding (SLU) is aimed at interpreting user’s intentions from their speech utterances. Traditionally, this has been accomplished by writing context-free grammars (CFGs) or unification grammars (UGs) manually. The manual grammar authoring process is laborious and expensive, requiring much expertise. In addition, robustness is a vital requirement of these modules, because the input of these modules comes from a speech recognition unit and always contains errors. In recent years, many data-driven models have been proposed for spoken language understanding, but these approaches have not been exploited in Persian dialog systems yet. The main purpose of this thesis is to implement the statistical approches in Persian language and modify them in order to handle Persian language complexities. For this purpose, an HMM-based understanding unit is implemented. Afterwards, a modified Viterbi algorithm was proposed to handle the word clusters in the Persian language. The performance of this algorithm is evaluated on the Peykare corpus. In addition, the two-step understanding model is used for a specific understanding task. Experimental results show that the statistical understanding methods are much more robust than the rule-based ones. Furthermore, the proposed modified Viterbi algorithm demonstrates a satisfactory performance in dealing with the word clusters in Persian and resulted in 27% reduction of POS tagging error.
- Keywords:
- Spoken Dialoge System ; Rule-Based Aproach ; Data Driven Method ; Spoken Langauge Understanding
Digital Object List-
محتواي پايان نامه - view
Bookmark