The impact of automated phonetic alignment and formant tracking workflows on sound change measurement
-
C. I. Watson
Abstract
In this study we investigated the impact that forced aligners and formant trackers have on sound change measurements. We compared the results from handchecked data to those where the process was automated. Data was taken from read passages in the Auckland Voices corpus of New Zealand English (NZE). There were 60 speakers: 33 under 25 years (17 women, 16 men) and 27 over 40 years of age (16 women and 11 men). Three forced aligners (MAUS, FAVE aligner, and Montreal Forced Aligner (MFA)) and three formant trackers (FOREST, SNACK, and PRAAT) were investigated in five different combinations (MAUS with FOREST, SNACK, and PRAAT and additionally FAVE and MFA with PRAAT). The formant values obtained using these processes were compared with formant data from hand-checked vowel tokens (both vowel boundaries and formant tracks). The hand-labelled data only used vowels with phrase stress. The study looked at the movements of the short front vowels in NZE and measured these changes using the vowel space measure. The results showed that there was significant variability introduced into an analysis of sound-change based on an automated forced-alignment and formant extraction workflow. As such, to support reproducibility of studies, it is critically important to include enough details about this workflow to avoid any ambiguity about the process.
Abstract
In this study we investigated the impact that forced aligners and formant trackers have on sound change measurements. We compared the results from handchecked data to those where the process was automated. Data was taken from read passages in the Auckland Voices corpus of New Zealand English (NZE). There were 60 speakers: 33 under 25 years (17 women, 16 men) and 27 over 40 years of age (16 women and 11 men). Three forced aligners (MAUS, FAVE aligner, and Montreal Forced Aligner (MFA)) and three formant trackers (FOREST, SNACK, and PRAAT) were investigated in five different combinations (MAUS with FOREST, SNACK, and PRAAT and additionally FAVE and MFA with PRAAT). The formant values obtained using these processes were compared with formant data from hand-checked vowel tokens (both vowel boundaries and formant tracks). The hand-labelled data only used vowels with phrase stress. The study looked at the movements of the short front vowels in NZE and measured these changes using the vowel space measure. The results showed that there was significant variability introduced into an analysis of sound-change based on an automated forced-alignment and formant extraction workflow. As such, to support reproducibility of studies, it is critically important to include enough details about this workflow to avoid any ambiguity about the process.
Chapters in this book
- Frontmatter I
- Contents V
- On the nature of speech dynamics: approaches to studying synchronic variation and diachronic change 1
-
Part 1: Empirical perspectives on diachronic change
- Fifty years of monophthong and diphthong shifts in Mainstream Australian English 17
- Coarticulation guides sound change: an acoustic-phonetic study of real-time change in word-initial /l/ over four decades of Glaswegian 49
- The impact of automated phonetic alignment and formant tracking workflows on sound change measurement 89
- One place, two speech communities: differing responses to sound change in Mainstream and Aboriginal Australian English in a small rural town 117
- Prosodic change in 100 years: the fall of the rise-fall in an Albanian variety 145
-
Part 2: Factors conditioning synchronic variation
- Control of larynx height in vowel production revisited: a real-time MRI study 175
- Sheila’s roses (are in the paddick): reduced vowels in Australian English 207
- The future of the queen: how to pronounce “König✶innen” ‘gender-neutrally’ in German 245
- Synchronic variation and diachronic change: mora-counting and syllable-counting dialects in Japanese 273
- Reconstructing the timeline of a consonantal change in a German dialect: evidence from agent-based modeling 307
-
Part 3: Theoretical approaches at the interface between synchronic variation and diachronic change
- On (mis)aligned innovative perception and production norms 343
- Phonological patterns and dependency relations may arise from aerodynamic factors 369
- Actuation without production bias 395
- Understanding the role of broadcast media in sound change 425
- Connecting prosody and duality of patterning in diachrony, typology, phylogeny, and ontogeny 453
- Index 483
Chapters in this book
- Frontmatter I
- Contents V
- On the nature of speech dynamics: approaches to studying synchronic variation and diachronic change 1
-
Part 1: Empirical perspectives on diachronic change
- Fifty years of monophthong and diphthong shifts in Mainstream Australian English 17
- Coarticulation guides sound change: an acoustic-phonetic study of real-time change in word-initial /l/ over four decades of Glaswegian 49
- The impact of automated phonetic alignment and formant tracking workflows on sound change measurement 89
- One place, two speech communities: differing responses to sound change in Mainstream and Aboriginal Australian English in a small rural town 117
- Prosodic change in 100 years: the fall of the rise-fall in an Albanian variety 145
-
Part 2: Factors conditioning synchronic variation
- Control of larynx height in vowel production revisited: a real-time MRI study 175
- Sheila’s roses (are in the paddick): reduced vowels in Australian English 207
- The future of the queen: how to pronounce “König✶innen” ‘gender-neutrally’ in German 245
- Synchronic variation and diachronic change: mora-counting and syllable-counting dialects in Japanese 273
- Reconstructing the timeline of a consonantal change in a German dialect: evidence from agent-based modeling 307
-
Part 3: Theoretical approaches at the interface between synchronic variation and diachronic change
- On (mis)aligned innovative perception and production norms 343
- Phonological patterns and dependency relations may arise from aerodynamic factors 369
- Actuation without production bias 395
- Understanding the role of broadcast media in sound change 425
- Connecting prosody and duality of patterning in diachrony, typology, phylogeny, and ontogeny 453
- Index 483
