Tampere University of Technology

TUTCRIS Research Portal

Automatic word count estimation from daylong child-centered recordings in various language environments using language-independent syllabification of speech

Research output: Contribution to journalArticleScientificpeer-review

Standard

Automatic word count estimation from daylong child-centered recordings in various language environments using language-independent syllabification of speech. / Räsänen, Okko; Seshadri, Shreyas; Karadayi, Julien; Riebling, Eric; Bunce, John; Cristia, Alejandrina; Metze, Florian; Casillas, Marisa; Rosemberg, Celia; Bergelson, Elika; Soderstrom, Melanie.

In: Speech Communication, Vol. 113, 01.10.2019, p. 63-80.

Research output: Contribution to journalArticleScientificpeer-review

Harvard

Räsänen, O, Seshadri, S, Karadayi, J, Riebling, E, Bunce, J, Cristia, A, Metze, F, Casillas, M, Rosemberg, C, Bergelson, E & Soderstrom, M 2019, 'Automatic word count estimation from daylong child-centered recordings in various language environments using language-independent syllabification of speech', Speech Communication, vol. 113, pp. 63-80. https://doi.org/10.1016/j.specom.2019.08.005

APA

Vancouver

Author

Räsänen, Okko ; Seshadri, Shreyas ; Karadayi, Julien ; Riebling, Eric ; Bunce, John ; Cristia, Alejandrina ; Metze, Florian ; Casillas, Marisa ; Rosemberg, Celia ; Bergelson, Elika ; Soderstrom, Melanie. / Automatic word count estimation from daylong child-centered recordings in various language environments using language-independent syllabification of speech. In: Speech Communication. 2019 ; Vol. 113. pp. 63-80.

Bibtex - Download

@article{889b650a014b42e497b30672f39f71cf,
title = "Automatic word count estimation from daylong child-centered recordings in various language environments using language-independent syllabification of speech",
abstract = "Automatic word count estimation (WCE) from audio recordings can be used to quantify the amount of verbal communication in a recording environment. One key application of WCE is to measure language input heard by infants and toddlers in their natural environments, as captured by daylong recordings from microphones worn by the infants. Although WCE is nearly trivial for high-quality signals in high-resource languages, daylong recordings are substantially more challenging due to the unconstrained acoustic environments and the presence of near- and far-field speech. Moreover, many use cases of interest involve languages for which reliable ASR systems or even well-defined lexicons are not available. A good WCE system should also perform similarly for low- and high-resource languages in order to enable unbiased comparisons across different cultures and environments. Unfortunately, the current state-of-the-art solution, the LENA system, is based on proprietary software and has only been optimized for American English, limiting its applicability. In this paper, we build on existing work on WCE and present the steps we have taken towards a freely available system for WCE that can be adapted to different languages or dialects with a limited amount of orthographically transcribed speech data. Our system is based on language-independent syllabification of speech, followed by a language-dependent mapping from syllable counts (and a number of other acoustic features) to the corresponding word count estimates. We evaluate our system on samples from daylong infant recordings from six different corpora consisting of several languages and socioeconomic environments, all manually annotated with the same protocol to allow direct comparison. We compare a number of alternative techniques for the two key components in our system: speech activity detection and automatic syllabification of speech. As a result, we show that our system can reach relatively consistent WCE accuracy across multiple corpora and languages (with some limitations). In addition, the system outperforms LENA on three of the four corpora consisting of different varieties of English. We also demonstrate how an automatic neural network-based syllabifier, when trained on multiple languages, generalizes well to novel languages beyond the training data, outperforming two previously proposed unsupervised syllabifiers as a feature extractor for WCE.",
keywords = "Automatic syllabification, Daylong recordings, Language acquisition, Noise robustness, Word count estimation",
author = "Okko R{\"a}s{\"a}nen and Shreyas Seshadri and Julien Karadayi and Eric Riebling and John Bunce and Alejandrina Cristia and Florian Metze and Marisa Casillas and Celia Rosemberg and Elika Bergelson and Melanie Soderstrom",
year = "2019",
month = "10",
day = "1",
doi = "10.1016/j.specom.2019.08.005",
language = "English",
volume = "113",
pages = "63--80",
journal = "Speech Communication",
issn = "0167-6393",
publisher = "Elsevier",

}

RIS (suitable for import to EndNote) - Download

TY - JOUR

T1 - Automatic word count estimation from daylong child-centered recordings in various language environments using language-independent syllabification of speech

AU - Räsänen, Okko

AU - Seshadri, Shreyas

AU - Karadayi, Julien

AU - Riebling, Eric

AU - Bunce, John

AU - Cristia, Alejandrina

AU - Metze, Florian

AU - Casillas, Marisa

AU - Rosemberg, Celia

AU - Bergelson, Elika

AU - Soderstrom, Melanie

PY - 2019/10/1

Y1 - 2019/10/1

N2 - Automatic word count estimation (WCE) from audio recordings can be used to quantify the amount of verbal communication in a recording environment. One key application of WCE is to measure language input heard by infants and toddlers in their natural environments, as captured by daylong recordings from microphones worn by the infants. Although WCE is nearly trivial for high-quality signals in high-resource languages, daylong recordings are substantially more challenging due to the unconstrained acoustic environments and the presence of near- and far-field speech. Moreover, many use cases of interest involve languages for which reliable ASR systems or even well-defined lexicons are not available. A good WCE system should also perform similarly for low- and high-resource languages in order to enable unbiased comparisons across different cultures and environments. Unfortunately, the current state-of-the-art solution, the LENA system, is based on proprietary software and has only been optimized for American English, limiting its applicability. In this paper, we build on existing work on WCE and present the steps we have taken towards a freely available system for WCE that can be adapted to different languages or dialects with a limited amount of orthographically transcribed speech data. Our system is based on language-independent syllabification of speech, followed by a language-dependent mapping from syllable counts (and a number of other acoustic features) to the corresponding word count estimates. We evaluate our system on samples from daylong infant recordings from six different corpora consisting of several languages and socioeconomic environments, all manually annotated with the same protocol to allow direct comparison. We compare a number of alternative techniques for the two key components in our system: speech activity detection and automatic syllabification of speech. As a result, we show that our system can reach relatively consistent WCE accuracy across multiple corpora and languages (with some limitations). In addition, the system outperforms LENA on three of the four corpora consisting of different varieties of English. We also demonstrate how an automatic neural network-based syllabifier, when trained on multiple languages, generalizes well to novel languages beyond the training data, outperforming two previously proposed unsupervised syllabifiers as a feature extractor for WCE.

AB - Automatic word count estimation (WCE) from audio recordings can be used to quantify the amount of verbal communication in a recording environment. One key application of WCE is to measure language input heard by infants and toddlers in their natural environments, as captured by daylong recordings from microphones worn by the infants. Although WCE is nearly trivial for high-quality signals in high-resource languages, daylong recordings are substantially more challenging due to the unconstrained acoustic environments and the presence of near- and far-field speech. Moreover, many use cases of interest involve languages for which reliable ASR systems or even well-defined lexicons are not available. A good WCE system should also perform similarly for low- and high-resource languages in order to enable unbiased comparisons across different cultures and environments. Unfortunately, the current state-of-the-art solution, the LENA system, is based on proprietary software and has only been optimized for American English, limiting its applicability. In this paper, we build on existing work on WCE and present the steps we have taken towards a freely available system for WCE that can be adapted to different languages or dialects with a limited amount of orthographically transcribed speech data. Our system is based on language-independent syllabification of speech, followed by a language-dependent mapping from syllable counts (and a number of other acoustic features) to the corresponding word count estimates. We evaluate our system on samples from daylong infant recordings from six different corpora consisting of several languages and socioeconomic environments, all manually annotated with the same protocol to allow direct comparison. We compare a number of alternative techniques for the two key components in our system: speech activity detection and automatic syllabification of speech. As a result, we show that our system can reach relatively consistent WCE accuracy across multiple corpora and languages (with some limitations). In addition, the system outperforms LENA on three of the four corpora consisting of different varieties of English. We also demonstrate how an automatic neural network-based syllabifier, when trained on multiple languages, generalizes well to novel languages beyond the training data, outperforming two previously proposed unsupervised syllabifiers as a feature extractor for WCE.

KW - Automatic syllabification

KW - Daylong recordings

KW - Language acquisition

KW - Noise robustness

KW - Word count estimation

U2 - 10.1016/j.specom.2019.08.005

DO - 10.1016/j.specom.2019.08.005

M3 - Article

VL - 113

SP - 63

EP - 80

JO - Speech Communication

JF - Speech Communication

SN - 0167-6393

ER -