Presented by

Sherif Abdou

VoiceXML experimentation platform

•Interpreter: SpeechWorks OpenVXI

•ASR engine: Lucent ASR (LASR)

•TTS engine: Lucent TTS

•Telephony interface : Dialogic Custom Telephony Interface (CTI)

Semantic Interpretation for Speech Recognition

$order = I would like a $drink {drink.liquid = $drink.type; drink.drinksize = $drink.drinksize} and $pizza {pizza=$pizza}; $kindofdrink = coke | pepsi | coca cola {"coke"};

$foodsize = {"medium"} [small | medium | large | regular {"medium"}];

$tops = $top {Append([],$top)} (and $top {Append($,$top)})<1->;

$top = anchovies | pepperoni | mushroom {"mushrooms"} | mushrooms;

$drink = $foodsize $kindofdrink {drinksize=$foodsize; type=$kindofdrink };

$pizza = $number $foodsize {pizzasize=$foodsize; number=$number} pizzas with $tops {topping=$tops}; 

$number = (a | one){1} | two {2}| three {3};

Speech recognition grammar

Recognized Utterance

"I would like a coca cola and three large pizzas with pepperoni and mushrooms."

Extracted semantic information

{ drink: { liquid:"coke" drinksize:"medium"} pizza: { number: "3" pizzasize: "large" topping: [ "pepperoni" "mushrooms" ] }}