Canada Spends Logo

Explorez notre analyse du budget 2026

Grant: Discovery Grants Program - Individual

University of Waterloo

$18,000.00

context-aware speech recognition

EF 2008-09

Résumé
The goal is to make an automatic speech recognizer that behaves more like the way people hear speech. Partly, this is done by forming groups of words that share a common feature, such as one sound in the word, so that whole groups can be tested at once based on this common feature. The way the sound is measured is also related more directly to movements of the mouth when people speak. Different measuring units are used for different types of sound, for example, vowels versus consonants. Since sounds occur in a different order in different words, this means that there may be different measuring units applied in a different order. Most speech recognizers use a common measuring unit for all sounds in all words so that interpretting the measurements is simpler but some distinctions between sounds might not show up as well. The grouping of words can also be based on other things, such as the way syllables are stressed or the part of grammar that describes the word, for example, a noun versus a verb. The desired outcome is to make it much easier to adapt the recognizer to different people or to different words in a vocabulary and possibly to reduce the number of mistakes made in recognition. Normally, the user must train a speech recognizer by reading out loud from a long story that contains examples of many different words. The proposed system should require much less of this training.
Programme
Discovery Grants Program - Individual
Emplacement
Ontario, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data

Tel que publié

Application id
6658-2007
Name nom
Freeman, George
Department département
Electrical & Computer Engineering
Organization id
33
Institution établissement
University of Waterloo
Province en
Ontario
Province fr
Ontario
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2008
Competition year année de concours
2007
Award amount
18000
Program id
RGPIN
Program nane en
Discovery Grants Program - Individual
Program name fr
Programme de subventions à la découverte - individuelles
Group en
Discovery grants programs
Group fr
Programmes de subventions à la découverte
Committee code
334
Committee name en
Commun.,Computers & Components
Committee name fr
Communic.,ordin. et composants
Area of application code
1207
Area of application group en
Advancement of knowledge
Area of application group fr
Avancement des connaissances
Area of application en
Information, computer and communication technologies
Area of application fr
Technologie de l'information, des ordinateurs et des réseaux de communications
Research subject code
2510
Research subject group en
Electrical and electronic engineering
Research subject group fr
Génie électrique et électronique
Research subject en
Adaptive, learning and evolutionary systems
Research subject fr
Systèmes d'apprentissage adaptatifs et évolutifs
Partie
2
Nb partie
5
Application title
context-aware speech recognition
Keyword
NA
Application summary
The goal is to make an automatic speech recognizer that behaves more like the way people hear speech. Partly, this is done by forming groups of words that share a common feature, such as one sound in the word, so that whole groups can be tested at once based on this common feature. The way the sound is measured is also related more directly to movements of the mouth when people speak. Different measuring units are used for different types of sound, for example, vowels versus consonants. Since sounds occur in a different order in different words, this means that there may be different measuring units applied in a different order. Most speech recognizers use a common measuring unit for all sounds in all words so that interpretting the measurements is simpler but some distinctions between sounds might not show up as well. The grouping of words can also be based on other things, such as the way syllables are stressed or the part of grammar that describes the word, for example, a noun versus a verb. The desired outcome is to make it much easier to adapt the recognizer to different people or to different words in a vocabulary and possibly to reduce the number of mistakes made in recognition. Normally, the user must train a speech recognizer by reading out loud from a long story that contains examples of many different words. The proposed system should require much less of this training.