Always private
DuckDuckGo never tracks your searches.
Learn More
You can hide this reminder in Search Settings
All regions
Argentina
Australia
Austria
Belgium (fr)
Belgium (nl)
Brazil
Bulgaria
Canada (en)
Canada (fr)
Catalonia
Chile
China
Colombia
Croatia
Czech Republic
Denmark
Estonia
Finland
France
Germany
Greece
Hong Kong
Hungary
Iceland
India (en)
Indonesia (en)
Ireland
Israel (en)
Italy
Japan
Korea
Latvia
Lithuania
Malaysia (en)
Mexico
Netherlands
New Zealand
Norway
Pakistan (en)
Peru
Philippines (en)
Poland
Portugal
Romania
Russia
Saudi Arabia
Singapore
Slovakia
Slovenia
South Africa
Spain (ca)
Spain (es)
Sweden
Switzerland (de)
Switzerland (fr)
Taiwan
Thailand (en)
Turkey
Ukraine
United Kingdom
US (English)
US (Spanish)
Vietnam (en)
Safe search: moderate
Strict
Moderate
Off
Any time
Any time
Past day
Past week
Past month
Past year
  1. sciencedirect.com

    Bottle-Neck Feature Extraction Structures for Multilingual ... Multilingual SBN training and porting The multilingual DNNs in SBN system are trained with the last layer â€" softmax â€" split into several blocks. ... This work was also supported by the European Union’s Horizon 2020 project No. 645523 BISON, and by Technology ...
  2. Was this helpful?
  3. researchgate.net

    Stacked-Bottle-Neck (SBN) feature extraction is a crucial part of modern automatic speech recognition (ASR) systems. The SBN network traditionally contains a hidden layer between the BN and output ...
  4. semanticscholar.org

    DOI: 10.1016/j.procs.2016.04.042 Corpus ID: 38091716; Bottle-Neck Feature Extraction Structures for Multilingual Training and Porting @inproceedings{Grzl2016BottleNeckFE, title={Bottle-Neck Feature Extraction Structures for Multilingual Training and Porting}, author={Frantisek Gr{\'e}zl and Martin Karafi{\'a}t}, booktitle={Workshop on Spoken Language Technologies for Under-resourced Languages ...
  5. mendeley.com

    In this paper, we describe two strategies allowing the direct-connection SBN network to indeed benefit from pre-training with a multilingual net: (1) pre-training multilingual net with the hidden layer which is discarded before porting to the target language and (2) using only the the direct- connection SBN with triphone targets both in ...
  6. apps.dtic.mil

    Keywords: Stacked Bottle-Neck; feature extraction; multilingual training; large data; Fisher database 1. Introduction Multilingual resources are of great help in case the data from the target language are not sufficient to train good acoustic model. In such case, the multilingual model, which is usually trained beforehand, is ported to the target
  7. isca-archive.org

    4.2. Multilingual bottle-neck feature We trained a multilingual MLP with English, French, German and Spanish training data. The MLP has 5 layers and has a topology 143-1500-42-1500-81. To make a comparison we trained also different monolingual MLPs with the same topol-ogy (only the number of target phones is changed). For all
  8. Stacked-Bottle-Neck (SBN) feature extraction is a crucial part of modern automatic speech recognition (ASR) systems. The SBN network traditionally contains a hidden layer between the BN and output layers. Recently, we have observed that an SBN architecture without this hidden layer (i.e. direct BN-layer - output-layer connection) performs better for a single language but fails in scenarios ...
  9. Office Hours. Mon 8:00-11:00 Wed 8:00-11:00 and 13:00-14:30 Thu 8:00-11:00 Note: during holidays from July to August office hours are only on Wednesdays 8:00-11:00. E-mail: study@fit.vut.cz Any e-mail communication should be addressed to study@fit.vut.cz.
  10. apps.dtic.mil

    Keywords: DNN topology; Stacked Bottle-Neck; feature extraction; multilingual training; system porting 1. Introduction One of the recent challenges in speech recognition community is to build an ASR system with limited in-domain data. The data hungry algorithms for training ASR system components have to be modified to be effective with less data.

    Can’t find what you’re looking for?

    Help us improve DuckDuckGo searches with your feedback

Custom date rangeX