This can be implemented by observing the traffic flow and increase the time of open signals where the flow of traffic is huge. This can be implemented with if else statements and saves life of people. Spoken language understanding:
Our basic concern in this paper is spoken language understanding in the scenario discussed above. For his purpose we will discuss different architectures for understanding spoken language and try to purpose an architecture that is fast as well as more effective. Problem statement:
• V.I.P movement
• Moving a prisoner from one place to another.
The trend of V.I.P movement is very common in our country. The instructions are given to the robot, the robot will understand the situation by the architecture implemented …show more content…
SUMMIT architecture:
This architecture has three basic pats or segments;
SUMMIT, TINA and control strategy:
SUMMIT converts speech signal into set of words.
TINA is the basic system as it interacts with the speech recognizer and generates a word string as a result. Other functionality of Tina area as following;

It is basically the grammar which is entered as a set of simple context-free rules which are automatically converted to a shared network structure of the system. The nodes in the network are handled with constraint filters working only on locally available parameters. Filters means handling both syntactic and semantic. All this developed arcs in the network are associated with probabilities, acquired automatically from a set of training sentences present in it.

Control Strategy: mediates between SUMMIT and TINA

LESK Algorithm:

This system Identify senses of words in context using definition overlap. Its algorithm includes;

Retrieve from MRD all sense definitions of the words to be disambiguated

Determine the definition overlap for all possible sense combinations
Choose senses that lead to highest …show more content…
It has following two steps
Step 1: For each sense of the target word find the thesaurus category to which that sense belongs.
Step 2: Calculate the score for each sense by using the context words. A context words will add 1 to the score of the sense if the thesaurus category of the word matches that of the sense.
E.g.; the money in this bank fetches an interest of 8% per annum
Target word: bank
Clue words from the context: money, interest, annum, fetches. PHOENIX Parser:
The Phoenix semantic parser is an easy parser which is used. This parser maps input word strings on to a sequence of semantic frames:
• Named set of slots, where the slots represent related pieces of information. • Each slot has an associated CFG that specifies word string patterns that match the slot
It is very simple and easy and provides very near and accurate results to the users.

Intent detection:

The input query which may be spoken or text is processed to extract features that convey the intent of information.

The intent detector employs a method similar to a likelihood ratio test to distinguish between two hypotheses: the input query is a known intent or not known. Score is computed against a model representing all other unknown (background)