Reliability of the Language Environment Analysis Recording System in Analyzing French-English Bilingual Speech

Item request has been placed! ×
Item request cannot be made. ×
loading   Processing Request
Read Online Read More Add to Saved list
  • Additional Information
    • Availability:
      American Speech-Language-Hearing Association. 2200 Research Blvd #250, Rockville, MD 20850. Tel: 301-296-5700; Fax: 301-296-8580; e-mail: [email protected]; Web site: http://jslhr.pubs.asha.org
    • Peer Reviewed:
      Y
    • Source:
      10
    • Subject Terms:
      French; Bilingualism; Infants; Linguistic Input; English; Audio Equipment; Parent Child Relationship; Language Usage; French Canadians; Speech Communication; Adults; Gender Differences; Pronunciation; English (Second Language); Second Language Learning; Language Acquisition; Word Frequency; Computational Linguistics; Comparative Analysis; Reliability; Foreign Countries; Language Variation
    • Subject Terms:
      Canada
    • Accession Number:
      10.1044/2019_JSLHR-L-18-0342
    • ISSN:
      1092-4388
    • Abstract:
      Purpose: This study examined the utility of the Language ENvironment Analysis (LENA) recording system for investigating the language input to bilingual infants. Method: Twenty-one French-English bilingual families with a 10-month-old infant participated in this study. Using the LENA recording system, each family contributed 3 full days of recordings within a 1-month period. A portion of these recordings (945 minutes) were manually transcribed, and the word counts from these transcriptions were compared against the LENA-generated adult word counts. Results: Data analyses reveal that the LENA algorithms were reliable in counting words in both Canadian English and Canadian French, even when both languages are present in the same recording. While the LENA system tended to underestimate the amount of speech in the recordings, there was a strong correlation between the LENA-generated and human-transcribed adult word counts for each language. Importantly, this relationship holds when accounting for different-gendered and different-accented speech. Conclusions: The LENA recording system is a reliable tool for estimating word counts, even for bilingual input. Special considerations and limitations for using the LENA recording system in a bilingual population are discussed. These results open up possibilities for investigating caregiver talk to bilingual infants in more detail.
    • Abstract:
      As Provided
    • Publication Date:
      2019
    • Accession Number:
      EJ1222516