Browsing by Subject "DROPOUT"

Sort by: Order: Results:

Now showing items 1-2 of 2
  • Smit, Peter; Virpioja, Sami; Kurimo, Mikko (2021)
    We describe a novel way to implement subword language models in speech recognition systems based on weighted finite state transducers, hidden Markov models, and deep neural networks. The acoustic models are built on graphemes in a way that no pronunciation dictionaries are needed, and they can be used together with any type of subword language model, including character models. The advantages of short subword units are good lexical coverage, reduced data sparsity, and avoiding vocabulary mismatches in adaptation. Moreover, constructing neural network language models (NNLMs) is more practical, because the input and output layers are small. We also propose methods for combining the benefits of different types of language model units by reconstructing and combining the recognition lattices. We present an extensive evaluation of various subword units on speech datasets of four languages: Finnish, Swedish, Arabic, and English. The results show that the benefits of short subwords are even more consistent with NNLMs than with traditional n-gram language models. Combination across different acoustic models and language models with various units improve the results further. For all the four datasets we obtain the best results published so far. Our approach performs well even for English, where the phoneme-based acoustic models and word-based language models typically dominate: The phoneme-based baseline performance can be reached and improved by 4% using graphemes only when several grapheme-based models are combined. Furthermore, combining both grapheme and phoneme models yields the state-of-the-art error rate of 15.9% for the MGB 2018 dev17b test. For all four languages we also show that the language models perform reasonably well when only limited training data is available.
  • Vesikivi, Petri; Lakkala, Minna; Holvikivi, Jaana; Muukkonen, Hanni (2020)
    Technological and social developments during the past years emphasise the importance of knowledge work competence. Additionally, funding of universities in Finland was changed to be based largely on yearly accumulated credits, therefore, improving retention is of critical importance for the institution. In order to improve first-year retention (measured by credit accumulation) and learning of knowledge work practices, Metropolia UAS changed the information technology curriculum by integrating single topic 3?5 credit courses into multidisciplinary 15 credit courses that included substantially more project work where students solve open-ended problems. This study focuses on investigating how the new curriculum influenced first-year retention, students? study experiences and self-evaluated development of knowledge work competence. Research data included study register data on course completion and student feedback collected through online questionnaires after each course. Retention rate was substantially improved compared to previous years. Furthermore, student collaboration and independence were found to increase overall satisfaction and to boost learning in project teams.