IERI Collaborative Research: Automating Early Assesment of Academic Standards for Very Young Native and Non-Native Speakers of American English

点击翻译按钮获取中文摘要

To help meet the increasing demand for high quality, efficient and accurate diagnostic assessments of children's academic skills, a new paradigm for automatic assessments is proposed. The project aims to advance the state of the art in speech processing, wireless communications, data mining, and human-computer interface (HCI) design so that effective child-friendly conversational interfaces can be designed and developed. These technologies will be researched in the framework of early learning and integrated with a progressive understanding of the components of academic performance to develop a literacy assessment system and explore the use of analogous assessment in math. The impact of the proposed approach will be studied with native speakers of American English (AE) and non-native AE speakers of Mexican-Spanish background in a longitudinal fashion starting from pre-K, and in partnership with the Los Angeles and Long Beach Unified School Districts, and UCLA's University Elementary School. These schools have a highly diverse economic and ethnic student body with more than half of the population being Hispanic. The project will analyze children's speech as they grow; develop speech recognition (ASR) algorithms for automating assessments that measure essential emerging literacy and some math skills; create a query-based, longitudinal database for each student; derive instructional guidance from the analysis of an ongoing professional development program for teachers of native and non-native speakers; and develop a nomadic interface among different computers and a central database. No feedback or tutoring will occur. Instead, teachers will use the results to make more timely and appropriate decisions about curriculum and instructional interventions. Technical Impact: The project will address several fundamental research issues: (a) acoustic modeling: documenting and accounting for inter- and intra-speaker variability cross-sectionally and longitudinally; (b) pronunciation modeling and speaker adaptation techniques that are scalable to children who are 4-8 years old and who are native and non-native English speakers; (c) child-specific language modeling: syntax, non-lexical events, and discourse phenomena; limited-domain natural-language processing (comprehension); (d) novel noise-robust and distributed ASR algorithms; (e) HCI: age-appropriate ways of displaying information and eliciting responses; (f) data mining: mining sequential patterns to discover trends over time, and user-specified associations; and (g) pedagogic issues: investigate early emerging literacy measures for native and non-native speakers, and discover reliable predictors of short- and long-term literacy success.Innovative aspects of the proposed approach include: a focus on literacy assessment that considers not only word recognition, but also phonetic and phonological awareness, comprehension and fluency; an exploratory study of automating math assessment tasks; a longitudinal and cross-sectional acoustic modeling study of very young native and non-native English speakers; extensive and longitudinal validation of the system, children's performance, and teachers' practices; correlating literacy measures to later reading performance; and pioneering research efforts involving system deployment in wired and wireless environments.Educational Impact: The project will foster interdisciplinary activities at: the U. of California, Los Angeles (Electrical Engineering [EE], Computer Science, and Education), U. of Southern California (EE, Linguistics, and Neuroscience), and U. of California, Berkeley (Education), in partnership with local elementary schools. Several renowned experts, including international experts from Mexico and Sweden, from academia and industry will act as advisory board members and consultants. Team members have a track record of working together, and the project will serve as a vehicle to train students, postdocs, and school teachers in novel cross-disciplinary research areas of technological and educational significance. Broader Impact: The proposed project will have a profound impact on relieving much of the burden of testing from teachers (allowing them to focus more on what they do best), automated testing for very young children (allowing a greater leverage point for potential intervention), and inclusion of an increasingly diverse population (enabling unbiased assessment and furthering the goal of universal access). National educational priorities are emphasizing testing to a greater extent than ever before, but increased testing leads to less time for teaching. The proposed system can reduce the test burden and increase the frequency of high-quality, intuitively consumable information about students so that individuals, programs, and schools can evolve more quickly by understanding which methods are working best for which children. Educational policy is also pushing downward to earlier ages to begin formal literacy instruction. This system will provide a useful aid to learn how to help young children succeed and to monitor their progress. The rapid expansion of student groups, reflecting diverse, non-native speakers of English, presents a challenge for fair assessment. The system helps ensure unbiased assessment of competence in a timely and useful way. It is expected that the project will have a profound impact on improving assessment and instructional material in the classroom.