Please use this identifier to cite or link to this item:
http://hdl.handle.net/10071/29215
Author(s): | Hämäläinen, A. Pinto, F. M. Rodrigues, S. Júdice, A. Silva, S. M. Calado, A. Dias, M. S. |
Editor: | Badin, P., Hueber, T., Bailly, G., Demolin, D., and Raby, F. |
Date: | 2013 |
Title: | A multimodal educational game for 3-10-year-old children: Collecting and automatically recognising European Portuguese children’s speech |
Book title/volume: | 2013 ISCA International Workshop on Speech and Language Technology in Education (SLaTE 2013) |
Pages: | 31 - 36 |
Event title: | 2013 ISCA International Workshop on Speech and Language Technology in Education (SLaTE 2013) |
Reference: | Hämäläinen, A., Pinto, F. M., Rodrigues, S., Júdice, A., Silva, S. M., Calado, A., & Dias, M. S. (2013). A multimodal educational game for 3-10-year-old children: Collecting and automatically recognising European Portuguese children’s speech. In P. Badin, T. Hueber, G. Bailly, D. Demolin, & F. Raby (Eds.), 2013 ISCA International Workshop on Speech and Language Technology in Education (SLaTE 2013) (pp. 31-36). The International Society for Computers and Their Applications (ISCA). https://www.isca-speech.org/archive/slate_2013/hamalainen13_slate.html |
ISSN: | 2311-4975 |
Keywords: | Acoustic modelling ASR Child-computer interaction Corpus Educational game European Portuguese |
Abstract: | Speech interfaces have tremendous potential in education. In this paper, we present our work in the Contents for Next Generation Networks project, an ongoing Portuguese industry-academia collaboration developing a multimodal educational game aimed at improving the physical coordination and the basic mathematical and musical skills of 3-10- year-old children. We focus on our work in the area of children's speech recognition: designing, collecting, transcribing and annotating a 21-hour corpus of prompted European Portuguese children's speech, as well as our first experiments with different acoustic modelling approaches. Our speech recognition results suggest that training children's speech models from scratch is a more promising approach than retraining adult speech models using children's speech when a sufficient amount of training data is available from the targeted age group. This finding also holds for adult female speech models retrained using children's speech. As compared with a baseline recogniser comprising gender-dependent adult speech models, the best-performing children's speech models that we have trained so far – genderindependent cross-word triphones trained with 17.5 hours of speech from 3-10-year-old children – resulted in a 45-percent (relative) decrease in word error rate in a task expecting isolated cardinal numbers, sequences of cardinal numbers or musical notes as speech input |
Peerreviewed: | yes |
Access type: | Open Access |
Appears in Collections: | ISTAR-CRI - Comunicações a conferências internacionais |
Files in This Item:
File | Size | Format | |
---|---|---|---|
conferenceobject_96466.pdf | 163,71 kB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.