Frans Adriaans
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Frans Adriaans has a total of 14 co-authors. Here are the top co-authors by number of shared publications:
- René Kager
Utrecht University
3 shared publications Jaap Kamps
University of Amsterdam
2 shared publications- Marijn Koolen
Royal Netherlands Academy of Arts and Sciences - Humanities Cluster, Netherlands
2 shared publications - Anika van der Klis
Experimental Psychology, Department of Psychology, Utrecht University, The Netherlands
1 shared publication - Ben Mark Harvey
Utrecht University
1 shared publication - Christian P. Janssen
Utrecht University, Utrecht, Netherlands
1 shared publication
Latest Publications
- Open Access
- 3 July 2023
- Published by MDPI AG in Multimodal Technologies and Interaction
- Vol. 7 (7) , 68
Using Open-Source Automatic Speech Recognition Tools for the Annotation of Dutch Infant-Directed Speech
There is a large interest in the annotation of speech addressed to infants. Infant-directed speech (IDS) has acoustic properties that might pose a challenge to automatic speech recognition (ASR) tools developed for adult-directed speech (ADS). While ASR tools could potentially speed up the annotation process, their effectiveness on this speech register is currently unknown. In this study, we assessed to what extent open-source ASR tools can successfully transcribe IDS. We used speech data from 21 Dutch mothers reading picture books containing target words to their 18- and 24-month-old children (IDS) and the experimenter (ADS). In Experiment 1, we examined how the ASR tool Kaldi-NL performs at annotating target words in IDS vs. ADS. We found that Kaldi-NL only found 55.8% of target words in IDS, while it annotated 66.8% correctly in ADS. In Experiment 2, we aimed to assess the difficulties in annotating IDS more broadly by transcribing all IDS utterances manually and comparing the word error rates (WERs) of two different ASR systems: Kaldi-NL and WhisperX. We found that WhisperX performs significantly better than Kaldi-NL. While there is much room for improvement, the results show that automatic transcriptions provide a promising starting point for researchers who have to transcribe a large amount of speech directed at infants.
- 3 July 2023
- Published by MDPI AG in Multimodal Technologies and Interaction
- Vol. 7 (7) , 68
Open AccessUsing Open-Source Automatic Speech Recognition Tools for the Annotation of Dutch Infant-Directed Speech
There is a large interest in the annotation of speech addressed to infants. Infant-directed speech (IDS) has acoustic properties that might pose a challenge to automatic speech recognition (ASR) tools developed for adult-directed speech (ADS). While ASR tools could potentially speed up the annotation process, their effectiveness on this speech register is currently unknown. In this study, we assessed to what extent open-source ASR tools can successfully transcribe IDS. We used speech data from 21 Dutch mothers reading picture books containing target words to their 18- and 24-month-old children (IDS) and the experimenter (ADS). In Experiment 1, we examined how the ASR tool Kaldi-NL performs at annotating target words in IDS vs. ADS. We found that Kaldi-NL only found 55.8% of target words in IDS, while it annotated 66.8% correctly in ADS. In Experiment 2, we aimed to assess the difficulties in annotating IDS more broadly by transcribing all IDS utterances manually and comparing the word error rates (WERs) of two different ASR systems: Kaldi-NL and WhisperX. We found that WhisperX performs significantly better than Kaldi-NL. While there is much room for improvement, the results show that automatic transcriptions provide a promising starting point for researchers who have to transcribe a large amount of speech directed at infants.
- 1 April 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Pervasive Computing
- Vol. 19 (2) , 63-68
Multidisciplinary and Interdisciplinary Teaching in the Utrecht AI Program: Why and How?
- 1 April 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Pervasive Computing
- Vol. 19 (2) , 63-68
Multidisciplinary and Interdisciplinary Teaching in the Utrecht AI Program: Why and How?
- Open Access
- 1 July 2018
- Published by Acoustical Society of America (ASA) in The Journal of the Acoustical Society of America
- Vol. 144 (1) , EL20-EL25
Effects of consonantal context on the learnability of vowel categories from infant-directed speech
Recent studies have shown that vowels in infant-directed speech (IDS) are characterized by highly variable formant distributions. The current study investigates whether vowel variability is partially due to consonantal context, and explores whether consonantal context could support the learning of vowel categories from IDS. A computational model is presented which selects contexts based on frequency in the input and generalizes across contextual categories. Improved categorization performance was found on a vowel contrast in American–English IDS. The findings support a view in which the infant's learning mechanism is anchored in context, in order to cope with acoustic variability in the input.
- 1 July 2018
- Published by Acoustical Society of America (ASA) in The Journal of the Acoustical Society of America
- Vol. 144 (1) , EL20-EL25
Open AccessEffects of consonantal context on the learnability of vowel categories from infant-directed speech
Recent studies have shown that vowels in infant-directed speech (IDS) are characterized by highly variable formant distributions. The current study investigates whether vowel variability is partially due to consonantal context, and explores whether consonantal context could support the learning of vowel categories from IDS. A computational model is presented which selects contexts based on frequency in the input and generalizes across contextual categories. Improved categorization performance was found on a vowel contrast in American–English IDS. The findings support a view in which the infant's learning mechanism is anchored in context, in order to cope with acoustic variability in the input.
- 26 July 2017
- Published by Wiley in Cognitive Science
- Vol. 42, 494-518
Linguistic Constraints on Statistical Word Segmentation: The Role of Consonants in Arabic and English
Statistical learning is often taken to lie at the heart of many cognitive tasks, including the acquisition of language. One particular task in which probabilistic models have achieved considerable success is the segmentation of speech into words. However, these models have mostly been tested against English data, and as a result little is known about how a statistical learning mechanism copes with input regularities that arise from the structural properties of different languages. This study focuses on statistical word segmentation in Arabic, a Semitic language in which words are built around consonantal roots. We hypothesize that segmentation in such languages is facilitated by tracking consonant distributions independently from intervening vowels. Previous studies have shown that human learners can track consonant probabilities across intervening vowels in artificial languages, but it is unknown to what extent this ability would be beneficial in the segmentation of natural language. We assessed the performance of a Bayesian segmentation model on English and Arabic, comparing consonant‐only representations with full representations. In addition, we examined to what extent structurally different proto‐lexicons reflect adult language. The results suggest that for a child learning a Semitic language, separating consonants from vowels is beneficial for segmentation. These findings indicate that probabilistic models require appropriate linguistic representations in order to effectively meet the challenges of language acquisition.
- 26 July 2017
- Published by Wiley in Cognitive Science
- Vol. 42, 494-518
Linguistic Constraints on Statistical Word Segmentation: The Role of Consonants in Arabic and English
Statistical learning is often taken to lie at the heart of many cognitive tasks, including the acquisition of language. One particular task in which probabilistic models have achieved considerable success is the segmentation of speech into words. However, these models have mostly been tested against English data, and as a result little is known about how a statistical learning mechanism copes with input regularities that arise from the structural properties of different languages. This study focuses on statistical word segmentation in Arabic, a Semitic language in which words are built around consonantal roots. We hypothesize that segmentation in such languages is facilitated by tracking consonant distributions independently from intervening vowels. Previous studies have shown that human learners can track consonant probabilities across intervening vowels in artificial languages, but it is unknown to what extent this ability would be beneficial in the segmentation of natural language. We assessed the performance of a Bayesian segmentation model on English and Arabic, comparing consonant‐only representations with full representations. In addition, we examined to what extent structurally different proto‐lexicons reflect adult language. The results suggest that for a child learning a Semitic language, separating consonants from vowels is beneficial for segmentation. These findings indicate that probabilistic models require appropriate linguistic representations in order to effectively meet the challenges of language acquisition.
- 1 May 2017
- Published by Acoustical Society of America (ASA) in The Journal of the Acoustical Society of America
- Vol. 141 (5) , 3070-3078
Prosodic exaggeration within infant-directed speech: Consequences for vowel learnability
Perceptual experiments with infants show that they adapt their perception of speech sounds toward the categories of the native language. How do infants learn these categories? For the most part, acoustic analyses of natural infant-directed speech have suggested that phonetic categories are not presented to learners as separable clusters of sounds in acoustic space. As a step toward explaining how infants begin to solve this problem, the current study proposes that the exaggerated prosody characteristic of infant-directed speech may highlight for infants certain speech-sound tokens that collectively form more readily identifiable categories. A database is presented, containing vowel measurements in a large sample of natural American English infant-directed speech. Analyses of the vowel space show that prosodic exaggeration in infant-directed speech has the potential to support distributional vowel learning by providing the learner with a subset of “high-quality” tokens that infants might attend to preferentially. Categorization models trained on prosodically exaggerated tokens outperformed models that were trained on tokens that were not exaggerated. Though focusing on more prominent, exaggerated tokens does not provide a solution to the categorization problem, it would make it easier to solve.
- 1 May 2017
- Published by Acoustical Society of America (ASA) in The Journal of the Acoustical Society of America
- Vol. 141 (5) , 3070-3078
Prosodic exaggeration within infant-directed speech: Consequences for vowel learnability
Perceptual experiments with infants show that they adapt their perception of speech sounds toward the categories of the native language. How do infants learn these categories? For the most part, acoustic analyses of natural infant-directed speech have suggested that phonetic categories are not presented to learners as separable clusters of sounds in acoustic space. As a step toward explaining how infants begin to solve this problem, the current study proposes that the exaggerated prosody characteristic of infant-directed speech may highlight for infants certain speech-sound tokens that collectively form more readily identifiable categories. A database is presented, containing vowel measurements in a large sample of natural American English infant-directed speech. Analyses of the vowel space show that prosodic exaggeration in infant-directed speech has the potential to support distributional vowel learning by providing the learner with a subset of “high-quality” tokens that infants might attend to preferentially. Categorization models trained on prosodically exaggerated tokens outperformed models that were trained on tokens that were not exaggerated. Though focusing on more prominent, exaggerated tokens does not provide a solution to the categorization problem, it would make it easier to solve.
- Open Access
- 5 January 2017
- Published by Open Library of the Humanities in Laboratory Phonology
- Vol. 8 (1) , 12
Learning novel phonotactics from exposure to continuous speech
Knowledge of phonotactics is commonly assumed to derive from the lexicon. However, computational studies have suggested that phonotactic constraints might arise before the lexicon is in place, in particular from co-occurrences in continuous speech. The current study presents two artificial language learning experiments aimed at testing whether phonotactic learning can take place in the absence of words. Dutch participants were presented with novel consonant constraints embedded in continuous artificial languages. Vowels occurred at random, which resulted in an absence of recurring word forms in the speech stream. In Experiment 1 participants with different training languages showed significantly different preferences on a set of novel test items. However, only one of the two languages resulted in preferences that were above chance-level performance. In Experiment 2 participants were exposed to a control language without novel statistical cues. Participants did not develop a preference for either phonotactic structure in the test items. An analysis of Dutch phonotactics indicated that the failure to induce novel phonotactics in one condition might have been due to interference from the native language. Our findings suggest that novel phonotactics can be learned from continuous speech, but participants have difficulty learning novel patterns that go against the native language
- 5 January 2017
- Published by Open Library of the Humanities in Laboratory Phonology
- Vol. 8 (1) , 12
Open AccessLearning novel phonotactics from exposure to continuous speech
Knowledge of phonotactics is commonly assumed to derive from the lexicon. However, computational studies have suggested that phonotactic constraints might arise before the lexicon is in place, in particular from co-occurrences in continuous speech. The current study presents two artificial language learning experiments aimed at testing whether phonotactic learning can take place in the absence of words. Dutch participants were presented with novel consonant constraints embedded in continuous artificial languages. Vowels occurred at random, which resulted in an absence of recurring word forms in the speech stream. In Experiment 1 participants with different training languages showed significantly different preferences on a set of novel test items. However, only one of the two languages resulted in preferences that were above chance-level performance. In Experiment 2 participants were exposed to a control language without novel statistical cues. Participants did not develop a preference for either phonotactic structure in the test items. An analysis of Dutch phonotactics indicated that the failure to induce novel phonotactics in one condition might have been due to interference from the native language. Our findings suggest that novel phonotactics can be learned from continuous speech, but participants have difficulty learning novel patterns that go against the native language
- 1 January 2012
- Published by Springer Nature in Lecture Notes in Computer Science
- p. 30-44
The Importance of Document Ranking and User-Generated Content for Faceted Search and Book Suggestions
- 1 January 2012
- Published by Springer Nature in Lecture Notes in Computer Science
- p. 30-44
The Importance of Document Ranking and User-Generated Content for Faceted Search and Book Suggestions
- 1 April 2010
- Published by Elsevier in Journal of Memory and Language
- Vol. 62 (3) , 311-331
Adding generalization to statistical learning: The induction of phonotactics from continuous speech
- 1 April 2010
- Published by Elsevier in Journal of Memory and Language
- Vol. 62 (3) , 311-331
Adding generalization to statistical learning: The induction of phonotactics from continuous speech
- 1 January 2006
- Published by Springer Nature
- p. 407-419
A Cross-Language Approach to Historic Document Retrieval
- 1 January 2006
- Published by Springer Nature
- p. 407-419
A Cross-Language Approach to Historic Document Retrieval