Canada Spends Logo

Explore our Budget 2026 analysis

Grant: Discovery Grants Program - Individual

The University of British Columbia

$40,000.00

Who is talking? What are they saying? The tangled web of recognizing talkers and understanding words

FY 2024-25

Summary
How do we understand the voice of someone we have never heard before? Humans possess a robust and reliable speech recognition system. In the face of infinite signal variability, adult listeners comprehend spoken language with apparent ease. Toddlers, non-native speakers, and computer-mediated speech are but a few examples of the disparate speech samples we regularly encounter and regularly comprehend. The speech patterns from these three sources are wildly distinct from one another due to a toddler's rapidly changing vocal tract anatomy and developing phonological knowledge, the colouring of a non-native speaker's speech patterns by their native language, and the signal processing of the computer. Now, how do we recognize the familiar voice of a mother, brother, or favourite neighbourhood barista? How do we learn the signatures of the voice of a new student? Learning and recognizing individuals' voices is a process that is distinct, but intertwined with spoken language recognition. A nearly ironclad empirical observation is that familiarity with the language being uttered helps listeners learn a novel voice. It is not clear, however, whether language knowledge provides a boost in the ability to access the relevant acoustic features that distinguish a particular novel voice, an advantage in learning a novel voice, or in one's ability to generalize the learned distinguishing features to novel utterances. The focus of the proposed research program is on spoken word recognition and voice organization in bilingual adults. The goals are multifold: (1) to better delineate shared and distinct processes between word and talker recognition; (2) to determine the use of language-specific versus language-general acoustic signatures in the organization of voices; (3) to identify processes and mechanisms that give some voices and listeners advantages in word and voice recognition. The focus on bilingual individuals is not only because this is the most common linguistic profile in Canada and the world, but also because this population introduces variability that our cognitive system can harness; we need to understand how. The methods of this work require detailed acoustic analysis, comparison of human and machine learning performance, and paradigms used in cognitive psychology and the language sciences. While our approach to these issues is from the perspective of understanding the mechanisms and processes of word recognition and voice organization, the research program informs myriad basic questions in speech perception and the structure and organization of linguistic knowledge. Beyond theory refinement and empirical discovery, this research has implications for users of automatic speech recognition (ASR) and speaker verification systems, particularly those whose language patterns push the boundaries of these systems (e.g., bilingual individuals who regularly mix their languages or whose accents are non-standard).
Program
Discovery Grants Program - Individual
Location
British Columbia, Canada
Recipient Type
Academia
Type
Grant
Source
NSERC's Awards Data

As published

Application id
RGPIN-2024-04989
Name nom
Babel, Molly
Department département
Linguistics
Organization id
2
Institution établissement
The University of British Columbia
Province en
British Columbia
Province fr
Colombie-Britannique
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2024
Competition year année de concours
2024
Award amount
40000
Program id
RGPIN
Program name en
Discovery Grants Program - Individual
Program name fr
Programme de subventions à la découverte - individuelles
Group en
DISCOVERY RESEARCH
Group fr
RECHERCHE AXÉE SUR LA DÉCOUVERTE
Committee code
1502
Committee name en
Biological Systems and Functions
Committee name fr
Systèmes et fonctions biologiques
Area of application code
1210
Area of application group en
Advancement of knowledge
Area of application group fr
Avancement des connaissances
Area of application en
Psychology
Area of application fr
Psychologie
Research subject code
6306
Research subject group en
Psychology
Research subject group fr
Psychologie
Research subject en
Cognitive science -- language
Research subject fr
Science cognitive -- langage
Field of research list names en
Not available
Field of research list names fr
Non disponible
Socioeconomic objective list names en
Not available
Socioeconomic objective list names fr
Non disponible
Application title
Who is talking? What are they saying? The tangled web of recognizing talkers and understanding words
Keywords
human voice; talker identification; spoken word recognition; machine learning; bilingualism
Application summary
How do we understand the voice of someone we have never heard before? Humans possess a robust and reliable speech recognition system. In the face of infinite signal variability, adult listeners comprehend spoken language with apparent ease. Toddlers, non-native speakers, and computer-mediated speech are but a few examples of the disparate speech samples we regularly encounter and regularly comprehend. The speech patterns from these three sources are wildly distinct from one another due to a toddler's rapidly changing vocal tract anatomy and developing phonological knowledge, the colouring of a non-native speaker's speech patterns by their native language, and the signal processing of the computer. Now, how do we recognize the familiar voice of a mother, brother, or favourite neighbourhood barista? How do we learn the signatures of the voice of a new student? Learning and recognizing individuals' voices is a process that is distinct, but intertwined with spoken language recognition. A nearly ironclad empirical observation is that familiarity with the language being uttered helps listeners learn a novel voice. It is not clear, however, whether language knowledge provides a boost in the ability to access the relevant acoustic features that distinguish a particular novel voice, an advantage in learning a novel voice, or in one's ability to generalize the learned distinguishing features to novel utterances. The focus of the proposed research program is on spoken word recognition and voice organization in bilingual adults. The goals are multifold: (1) to better delineate shared and distinct processes between word and talker recognition; (2) to determine the use of language-specific versus language-general acoustic signatures in the organization of voices; (3) to identify processes and mechanisms that give some voices and listeners advantages in word and voice recognition. The focus on bilingual individuals is not only because this is the most common linguistic profile in Canada and the world, but also because this population introduces variability that our cognitive system can harness; we need to understand how. The methods of this work require detailed acoustic analysis, comparison of human and machine learning performance, and paradigms used in cognitive psychology and the language sciences. While our approach to these issues is from the perspective of understanding the mechanisms and processes of word recognition and voice organization, the research program informs myriad basic questions in speech perception and the structure and organization of linguistic knowledge. Beyond theory refinement and empirical discovery, this research has implications for users of automatic speech recognition (ASR) and speaker verification systems, particularly those whose language patterns push the boundaries of these systems (e.g., bilingual individuals who regularly mix their languages or whose accents are non-standard).
The University of British Columbia - $40,000 Grant (2… | Canada Spends