Home	Company	News	Product	Partners	R & D	Support	Contact	فارسی

Home > R & D > Projects

	R & D  Overview         Projects          People      Publication           Contact with ASR Gooyesh Pardaz experts 2nd floor, No. 1, Teymoori alley, Shahid ghasemi St. Azadi. Av. Back of Sharif university of Technology (R&D manager) Tel.-Fax: (98) 21 66003710 Or contact via rd@asr-gooyesh.com	Projects of R&D group in ASR Gooyesh Pardaz Some of the current projects in the ASR Gooyesh Pardaz' research group are as follows. Also the publications of these projects are available at Publication page.   Speech Recognition: Dictation system NEVISA is a Persian dictation system developed as a result of this project. In fact NEVISA is the recognition engine which uses most popular algorithms and methods in the speech recognition field. This engine uses HMM-based modeling and MFCC as a core feature extraction with some modifications.   Robust Speech Recognition This project has been started since two years ago and now has developed many of the approaches to noise and speaker robustness. Since our goal is to develop speech recognition related applications operational in real environments, many of these methods have been developed and finalized in the recognition engine. Some of these methods are: On robust features: CMS, PCA, RASTA-PLP, RCC, Liftering On speech enhancement: Spectral Subtraction, Microphone array and beam-forming On model adaptation: MLLR and MAP On model prediction: PMC On speaker normalization: VTLN   Language modeling and Natural Language Processing For any spoken language related systems, linguistic information is the necessary part of that system. For the first time in Persian language, statistical and grammatical language models have been prepared and developed by ASR Gooyesh Pardaz' research group. Also equivalent researches are performing for English language and our English speech recognition engine. Some of the prepared linguistic information are: N-Grams (N=1,2,3) for Persian and English Grammatical rules using GPSG for Persian Probabilistic grammars Word clustering   Telephony Speech Recognition NEWSHA is the first speech-enabled Persian computer-telephony system. It is the result of more than three years research in this group.   Embedded Speech Recognition One of our missions in SPL is to develop an embedded speech recognition engine on low resource computers like smart phones and PDAs. Voice translator and Application launcher are our two primary applications in this area.   Keyword Spotting Keyword spotting, finding specific words in an audio stream, is another research field in SPL. The first version of this software is now available in Persian and English.   Confidence measure and Out Of Vocabulary Ranking the recognized word or word sequence is a necessary ability in speech recognition and word spotting systems. In order to have a practical system this ranking and detecting the out of vocabulary words are vital especially in command recognition systems. This projects has been active since two years ago.   Speech Enhancement Another field in our research is in Speech Quality Enhancement. Spectral Subtraction and Wiener Filter as two classic methods was experienced and some other approaches like signal sub-space and array processing beam-forming are in progress.   Voice Activity Detection (VAD) For detecting voice signals from non-voice ones, each speech-based system and specially recognition and enhancement systems needs a VAD block. Here we worked on VAD standards, ETSI's AMR and ITU-T's G.722 VAD and also developed two new other ones. These VADs are now incorporated in our recognition engine.   Distance talking and microphone array Distance talking and speaker localization are the main topics of this project.   Speech synthesis: Text-To-Speech (TTS) SPL also researches on TTS methods and tries to develop practical synthesis system in order to incorporate into other applications like telephony speech-enabled systems.   Native and non-native pronunciation ranking Ranking the pronunciation of an utterance is one of the major parts of language learning applications. We have used different approaches, especially HMM-based ranking, to achieve this goal.   Fast likelihood computation Likelihood computation is one of the main obstacles to move a speech recognition engine to low-resource computers and devices. Hence various fast Likelihood computation approaches are implemented in order to decrease the computational load in real-time applications.

About Us | Contact Us | Site Map

Copyright © 2002-2007 ASR Gooyesh Pardaz Co. All Rights Reserved