CHILDES Japanese MiiPro Corpus

Susanne Miyata Department of Medical Sciences Aichi Shukotoku University smiyata@asu.aasa.ac.jp website

Hiro Yuki Nisisawa College of Human Science Tokiwa University nisisawa@tokiwa.ac.jp website

Browsable transcripts

Download transcripts

Link to media folder

Citation information

Miyata, S. (2012a). Nihongo MLU (heikin hatsuwachō) no gaidorain: Jiritsugo MLU oyobi keitaiso MLU no keisanhō [Guideline for Japanese MLU: How to compute MLUw and MLUm]. Kenkō Iryō Kagaku 2, 1–15. http://aska-r.aasa.ac.jp/dspace/bitstream/10638/5113/1/0039-002-201203-1 -17.pdf (accessed 5 November 2012).

Miyata, S. (2012b). CHILDES nihongoban: Nihongoyoo CHILDES manyuaru 2012. [Japanese CHILDES: The 2012 CHILDES manual for Japanese]. http://www2.aasa.ac.jp/people/smiyata/CHILDESmanual/chapter01.html

Miyata, S. & Naka, N. (2014). JMOR06.2: The Japanese morphological analysis program based on CLAN. available at:

Oshima-Takane, Y. & B. MacWhinney (eds.) (1995, 2nd ed. 1998). CHILDES Manual for Japanese. Montreal: McGill University / Nagoya: Chukyo University.

In accordance with TalkBank rules, any use of data from this corpus must be accompanied by at least one of the above references.

Project Description

Miyata, Susanne smiyata@asu.aasa.ac.jp

Nisisawa, Hiro Yuki nisisawa@tokiwa.ac.jp

The "MiiPro" study observes 4 acquainted children of the same age, living close to each other in a neighborhood in the Tokyo area. Each child was observed separately at his/her home on a weekly basis from 1;2 to 3;0, and monthly or bi-monthly from 3;0 to 5;0. Each mother-child session lasted approx. 70 minutes.

The "MiiPro" study observes 4 acquainted children of the same age, living close to each other in a neighborhood in the Tokyo area. Each child was observed separately at his/her home on a weekly basis from 1;2 to 3;0, and monthly or bi-monthly from 3;0 to 5;0. Each mother-child session lasted approx. 70 minutes. Details about the files are available here .

Tomito Corpus

Miyata, Susanne smiyata@asu.aasa.ac.jp

Nisisawa, Hiro Yuki nisisawa@tokiwa.ac.jp

The boy Tomito (1;2.10 – 5;1.23) was observed monthly for approx. 70 minute. The corpus includes 61 sessions with a total of 29,887 utterances of Tomito and 73,571 utterances of MOT. Tomito’s younger sister Honoka was present and sometimes took part in the conversation. Note that only the utterances of Honoka that were related to Tomito’s conversation with their mother, have been transcribed. Honoka’s transcribed utterances are not in JCHAT format and have not been checked for reliability.

Asato Corpus

Miyata, Susanne smiyata@asu.aasa.ac.jp

Nisisawa, Hiro Yuki nisisawa@tokiwa.ac.jp

The first-born boy Asato (1;1.27 – 5;0.27) was observed weekly up to 3;0. Between 3;0 and 3;9 he was observed monthly, and after a 3-months break every second month (each session approx. 70 minutes; total time 74h). The corpus (1;2 – 5;1) includes 63 sessions with a total of 30,475 utterances from Asato and 54,013 utterances from his mother.

Nanami Corpus

Nisisawa, Hiro Yuki nisisawa@tokiwa.ac.jp

Miyata, Susanne smiyata@asu.aasa.ac.jp

The girl Nanami (1;1.29 – 5;0.17) was observed monthly. The corpus includes 57 sessions with a total of 27,912 utterances of Nanami and 60,118 utterances of MOT. Nanami's younger sister Juri (JUI; born when Nanami was 2;2) and her baby brother Kanta (KAN; born when Nanami was 3;6) were mostly present and sometimes took part in the conversation. Their transcribed utterances are not in JCHAT format and have not been checked for reliability.

Arika Corpus

Nisisawa, Hiro Yuki nisisawa@tokiwa.ac.jp

Miyata, Susanne smiyata@asu.aasa.ac.jp

The girl Arika (1;2.30 – 5;0.17) was observed weekly up to 3;0, and afterwards monthly. The Arika corpus includes 109 sessions with a total of 74,266 utterances of Arika and 80,138 utterances of MOT. Arika's younger sister Fuyumi (FUY; born when Arika was 3;2) was often present and started to take part in the conversation in the later sessions. Fuyumi’s transcribed utterances are not in JCHAT format and have not been checked for reliability