Projects funded by the NCN


Information on the principal investigator and host institution

Information of the project and the call

Keywords

Equipment

Delete all

Applications of deep and recurrent neural networks in acoustic modeling of speech

2014/15/D/ST6/04402

Keywords:

neural networks deep learning recurrent neural networks unsupervised pretraining representation learning

Descriptors:

  • ST6_11: Machine learning, statistical data processing and applications using signal processing (e.g. speech, image, video)
  • ST6_7: Artificial intelligence, intelligent systems, multi-agent systems
  • ST6_9: Human computer interaction, speech recognition and synthesis, natural language processing

Panel:

ST6 - Computer science and informatics: informatics and information systems, computer science, scientific computing, intelligent systems

Host institution :

Uniwersytet Wrocławski, Wydział Matematyki i Informatyki

woj. dolnośląskie

Other projects carried out by the institution 

Principal investigator (from the host institution):

dr Jan Chorowski 

Number of co-investigators in the project: 3

Call: SONATA 8 - announced on 2014-09-15

Amount awarded: 307 550 PLN

Project start date (Y-m-d): 2015-07-22

Project end date (Y-m-d): 2019-07-21

Project duration:: 48 months (the same as in the proposal)

Project status: Project settled

Equipment purchased [PL]

  1. Stanowisko obliczeniowe (30 000 PLN)

Information in the final report

  • Articles in post-conference publications (10)
  • Book publications / chapters in book publications (1)
  1. End-to-end attention-based large vocabulary speech recognition
    Authors:
    Dzmitry Bahdanau, Jan Chorowski, Dmitriy Serdyuk, Philemon Brakel, Yoshua Bengio
    Conference:
    Acoustics, Speech and Signal Processing (ICASSP) (rok: 2016, ), Wydawca: IEEE
    Data:
    konferencja 20-25.3.2016
    Status:
    Published
  2. Robust Training of Vector Quantized Bottleneck Models
    Authors:
    Adrian Łańcucki; Jan Chorowski; Guillaume Sanchez; Ricard Marxer; Nanxin Chen; Hans J.G.A. Dolfing; Sameer Khurana; Tanel Alumäe; Antoine Laurent
    Conference:
    2020 International Joint Conference on Neural Networks (IJCNN) (rok: 2020, ), Wydawca: IEEE
    Data:
    konferencja 19-24.07.2020
    Status:
    Published
  3. Towards Using Context-Dependent Symbols in CTC Without State-Tying Decision Trees
    Authors:
    Jan Chorowski, Adrian Łańcucki, Bartosz Kostka, and Michał Zapotoczny
    Conference:
    Interspeech 2019 (rok: 2019, ), Wydawca: ISCA
    Data:
    konferencja Wrzesień
    Status:
    Published
  4. Lattice Generation in Attention- Based Speech Recognition Models
    Authors:
    Michał Zapotoczny, Piotr Pietrzak, Adrian Łańcucki, and Jan Chorowski
    Conference:
    Interspeech 2019 (rok: 2019, ), Wydawca: ISCA
    Data:
    konferencja Wrzesień
    Status:
    Published
  5. Efficient Purely Convolutional Text Encoding
    Authors:
    Szymon Malik, Adrian Lancucki, Jan Chorowski
    Conference:
    Warsztaty LaCATODA przy ICML 2018 (rok: 2018, ), Wydawca: CEUR
    Data:
    konferencja 13.07.2018
    Status:
    Published
  6. Multipoint Neighbor Embedding
    Authors:
    Adrian Lancucki, Jan Chorowski
    Conference:
    Text, Speech, and Dialogue (rok: 2017, ), Wydawca: Springer
    Data:
    konferencja August 27-31
    Status:
    Published
  7. Unsupervised Neural Segmentation and Clustering for Unit Discovery in Sequential Data
    Authors:
    Jan K Chorowski, Nanxin Chen, Ricard Marxer, Hans J.G.A. Dolfing, Adrian Łańcucki, Guillaume Sanchez, Sameer Khurana, Tanel Alumäe, Antoine Laurent
    Conference:
    NIPS 2019 Workshop on Perception as Generative Reasoning (rok: 2019, ), Wydawca: Workshop organizers
    Data:
    konferencja 13.12.2019
    Status:
    Published
  8. Character-Based Neural POS Tagger
    Authors:
    Paweł Rychlikowski, Michał Zapotoczny and Jan Chorowski
    Conference:
    Language & Technology Conference (rok: 2017, ), Wydawca: Language & Technology Conference
    Data:
    konferencja November 17-19
    Status:
    Published
  9. Attention-based models for speech recognition
    Authors:
    Jan K Chorowski, Dzmitry Bahdanau, Dmitriy Serdyuk, Kyunghyun Cho, Yoshua Bengio
    Conference:
    Neural Information Processing Systems (rok: 2015, ), Wydawca: NIPS Foundation
    Data:
    konferencja 7-012.12.2015
    Status:
    Published
  10. On Multilingual Training of Neural Dependency Parsers.
    Authors:
    Michal Zapotoczny, Pawel Rychlikowski, Jan Chorowski
    Conference:
    Text, Speech, and Dialogue 2017 (rok: 2017, ), Wydawca: Springer
    Data:
    konferencja August 27-31
    Status:
    Published
  1. A Talker Ensemble: The University of Wroclaw's Entry to the NIPS 2017 Conversational Intelligence Challenge
    Authors:
    Jan ChorowskiAdrian LancuckiSzymon MalikMaciej PawlikowskiPawel RychlikowskiPawel Zykowski
    Book:
    The NIPS '17 Competition: Building Intelligent Systems (rok: 2018, tom: 1, strony: 59-77), Wydawca: Springer
    Status:
    Published