Canada Spends Logo

Explore our Budget 2026 analysis

Grant: Discovery Grants Program - Individual

McGill University

$60,545.00

Explaining sound change through mapping the distribution of phonetic biases and automatic speech analysis

FY 2023-24

Summary
A fundamental aspect of speech is massive variability in how the "same sound" is pronounced: across languages, speakers, speech styles, and over time. The long-term goal of this research program is to understand the structure of variability in speech production, enabled by novel quantitative methods for studying speech variability. This project addresses one type of speech variability: change over (historical) time, or sound change, which arise from small "phonetic effects" which become exaggerated over time. For example, tone languages (e.g. Mandarin), where large pitch differences in vowels ("tones") differentiate words, arise historically from small pitch differences in vowels. While what sound changes occur in human languages have been well catalogued we are a long way from understanding why: which phonetic effects can (and cannot) lead to sound change, and what role is played by differences between people (in their accents). This is largely because exists because the empirical data to develop and test such theories does not exist, namely the detailed distributions of phonetic effects across languages and speakers. Mapping these distributions in large speech datasets is our focus, through three short-term objectives: ?Objective 1: Establish the detailed cross-linguistic distribution of phonetic effects across many languages. ?Objective 2: Establish the distribution of variability in phonetic effects across many speakers. ?Objective 3: Enable large-scale studies of speech variability through automatic analysis tools The results of this interdisciplinary project will be of interest to researchers in linguistics, speech science, cognitive science, and computer science, as well as non-academic audiences. The description of phonetic effects cross-linguistically will provide important data for researchers developing theories of language change and speech production. The project contributes to a methodological shift in speech/language research towards larger and more naturalistic sources of data, and provides tools to work more easily with such data. By helping to accelerate research on disordered speech, these tools could help lead to improved treatments for speech and communication disorders. The project will result in large public speech datasets for Canadian English and French, which could spur new technologies (e.g. more natural speech synthesis in these dialects). It will help train the next generation of language scientists and data scientists, both important and growing fields, in quantitative and communication skills which are crucial for academic careers and highly transferable to non-academic careers.
Program
Discovery Grants Program - Individual
Location
Quebec, Canada
Recipient Type
Academia
Type
Grant
Source
NSERC's Awards Data

As published

Application id
RGPIN-2023-04873
Name nom
Sonderegger, Morgan
Department département
Linguistics - Linguistique
Organization id
61
Institution établissement
McGill University
Province en
Québec
Province fr
Québec
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2023
Competition year année de concours
2023
Award amount
60545
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
1200
Area of application group en
Advancement of knowledge
Area of application group fr
Avancement des connaissances
Area of application en
Advancement of knowledge
Area of application fr
Avancement des connaissances
Research subject code
2801
Research subject group en
Artificial intelligence
Research subject group fr
Natural language and speech understanding
Research subject en
Intelligence artificielle
Research subject fr
Langage naturel et reconnaissance de la parole
Application title
Explaining sound change through mapping the distribution of phonetic biases and automatic speech analysis
Keyword
computational linguistics; individual differences; language variation; linguistic corpora; phonetics; phonology; quantitative methods; sound change; speech production; speech technology
Application summary
A fundamental aspect of speech is massive variability in how the "same sound" is pronounced: across languages, speakers, speech styles, and over time. The long-term goal of this research program is to understand the structure of variability in speech production, enabled by novel quantitative methods for studying speech variability. This project addresses one type of speech variability: change over (historical) time, or sound change, which arise from small "phonetic effects" which become exaggerated over time. For example, tone languages (e.g. Mandarin), where large pitch differences in vowels ("tones") differentiate words, arise historically from small pitch differences in vowels. While what sound changes occur in human languages have been well catalogued we are a long way from understanding why: which phonetic effects can (and cannot) lead to sound change, and what role is played by differences between people (in their accents). This is largely because exists because the empirical data to develop and test such theories does not exist, namely the detailed distributions of phonetic effects across languages and speakers. Mapping these distributions in large speech datasets is our focus, through three short-term objectives: ?Objective 1: Establish the detailed cross-linguistic distribution of phonetic effects across many languages. ?Objective 2: Establish the distribution of variability in phonetic effects across many speakers. ?Objective 3: Enable large-scale studies of speech variability through automatic analysis tools The results of this interdisciplinary project will be of interest to researchers in linguistics, speech science, cognitive science, and computer science, as well as non-academic audiences. The description of phonetic effects cross-linguistically will provide important data for researchers developing theories of language change and speech production. The project contributes to a methodological shift in speech/language research towards larger and more naturalistic sources of data, and provides tools to work more easily with such data. By helping to accelerate research on disordered speech, these tools could help lead to improved treatments for speech and communication disorders. The project will result in large public speech datasets for Canadian English and French, which could spur new technologies (e.g. more natural speech synthesis in these dialects). It will help train the next generation of language scientists and data scientists, both important and growing fields, in quantitative and communication skills which are crucial for academic careers and highly transferable to non-academic careers.