Recent Submissions

  • Towards mitigating gender bias in a decoder-based neural machine translation model by adding contextual information 

    Basta, Christine Raouf Saad; Ruiz Costa-Jussà, Marta; Rodríguez Fonollosa, José Adrián (Association for Computational Linguistics, 2020)
    Conference report
    Open Access
    Gender bias negatively impacts many natural language processing applications, including ma-chine translation (MT). The motivation behind this work is to study whether recent proposedMT techniques are significantly contributing ...
  • Unsupervised training of siamese networks for speaker verification 

    Khan, Umair; Hernando Pericás, Francisco Javier (International Speech Communication Association (ISCA), 2020)
    Conference report
    Open Access
    Speaker labeled background data is an essential requirement for most state-of-the-art approaches in speaker recognition, e.g., xvectors and i-vector/PLDA. However, in reality it is difficult to access large amount of labeled ...
  • Combining subword representations into word-level representations in the transformer architecture 

    Casas Manzanares, Noé; Ruiz Costa-Jussà, Marta; Rodríguez Fonollosa, José Adrián (Association for Computational Linguistics, 2020)
    Conference lecture
    Open Access
    In Neural Machine Translation, using word-level tokens leads to degradation in translation quality. The dominant approaches use subword-level tokens, but this increases the length of the sequences and makes it difficult ...
  • Automatic Spanish translation of SQuAD dataset for multi-lingual question answering 

    Carrino, Casimiro Pio; Ruiz Costa-Jussà, Marta; Rodríguez Fonollosa, José Adrián (European Language Resources Association (ELRA), 2020)
    Conference lecture
    Open Access
    Recently, multilingual question answering became a crucial research topic, and it is receiving increased interest in the NLP community.However, the unavailability of large-scale datasets makes it challenging to train ...
  • Evaluating the underlying gender bias in contextualized word embeddings 

    Basta, Christine Raouf Saad; Ruiz Costa-Jussà, Marta; Casas Manzanares, Noé (Association for Computational Linguistics, 2019)
    Conference report
    Open Access
    Gender bias is highly impacting natural language processing applications. Word embeddings have clearly been proven both to keep and amplify gender biases that are present in current data sources. Recently, contextualized ...
  • I-vector transformation using k-nearest neighbors for speaker verification 

    Khan, Umair; India Massana, Miquel Àngel; Hernando Pericás, Francisco Javier (Institute of Electrical and Electronics Engineers (IEEE), 2020)
    Conference report
    Restricted access - publisher's policy
    Probabilistic Linear Discriminant Analysis (PLDA) is the most efficient backend for i-vectors. However, it requires labeled background data which can be difficult to access in practice. Unlike PLDA, cosine scoring avoids ...
  • From bilingual to multilingual neural machine translation by incremental training 

    Escolano Peinado, Carlos; Ruiz Costa-Jussà, Marta; Rodríguez Fonollosa, José Adrián (Association for Computational Linguistics, 2019)
    Conference lecture
    Open Access
    Multilingual Neural Machine Translation approaches are based on the use of task specific models and the addition of one more language can only be done by retraining the whole system. In this work, we propose a new training ...
  • Multilingual, multi-scale and multi-layer visualization of sequence-based intermediate representations 

    Escolano Peinado, Carlos; Ruiz Costa-Jussà, Marta; Lacroux, Elora; Vázquez Alcocer, Pere Pau (Association for Computational Linguistics, 2019)
    Conference report
    Restricted access - publisher's policy
    The main alternatives nowadays to dealwith sequences are Recurrent Neural Net-works (RNN), Convolutional Neural Networks(CNN) architectures and the Transformer. Inthis context, RNN’s, CNN’s and Transformerhave most commonly ...
  • Self multi-head attention for speaker recognition 

    India Massana, Miquel Àngel; Safari, Pooyan; Hernando Pericás, Francisco Javier (International Speech Communication Association (ISCA), 2019)
    Conference lecture
    Open Access
    Most state-of-the-art Deep Learning (DL) approaches forspeaker recognition work on a short utterance level. Given thespeech signal, these algorithms extract a sequence of speakerembeddings from short segments and those are ...
  • Auto-encoding nearest neighbor i-vectors for speaker verification 

    Khan, Umair; India Massana, Miquel Àngel; Hernando Pericás, Francisco Javier (International Speech Communication Association (ISCA), 2019)
    Conference lecture
    Open Access
    In the last years, i-vectors followed by cosine or PLDA scoringtechniques were the state-of-the-art approach in speaker veri-fication. PLDA requires labeled background data, and thereexists a significant performance gap ...
  • DNN speaker embeddings using autoencoder pre-training 

    Khan, Umair; Hernando Pericás, Francisco Javier (Institute of Electrical and Electronics Engineers (IEEE), 2019)
    Conference lecture
    Restricted access - publisher's policy
    Over the last years, i-vectors have been the state-of-the-art approach in speaker recognition. Recent improvements in deep learning have increased the discriminative quality of i-vectors. However, deep learning architectures ...
  • The TALP-UPC machine translation systems for WMT19 news translation task: pivoting techniques for low resource MT 

    Casas Manzanares, Noé; Rodríguez Fonollosa, José Adrián; Escolano Peinado, Carlos; Basta, Christine Raouf Saad; Ruiz Costa-Jussà, Marta (Association for Computational Linguistics, 2019)
    Conference report
    Restricted access - publisher's policy
    In this article, we describe the TALP-UPC research group participation in the WMT19 news translation shared task for Kazakh-English. Given the low amount of parallel training data, we resort to using Russian as pivot ...

View more