Výsledky vyhledávání - "Daniel Soutner"

Recurrent Neural Network Based Speaker Change Detection from Text Transcription Applied in Telephone Speaker Diarization System

Autor: Luděk Müller, Vlasta Radová, Daniel Soutner, Zbyněk Zajíc, Marek Hrúz

Publikováno v: Text, Speech, and Dialogue ISBN: 9783030007935
TSD

In this paper, we propose a speaker change detection system based on lexical information from the transcribed speech. For this purpose, we applied a recurrent neural network to decide if there is an end of an utterance at the end of a spoken word. Ou

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::46cd4b62bfac7c2c42a8bae9ab4ee545
https://doi.org/10.1007/978-3-030-00794-2_37

Zobrazit plný text záznamu

A Regularization Post Layer: An Additional Way How to Make Deep Neural Networks Robust

Autor: Josef Psutka, Jan Zelinka, Daniel Soutner, Jan Vaněk

Publikováno v: Statistical Language and Speech Processing ISBN: 9783319684550
SLSP

Neural Networks (NNs) are prone to overfitting. Especially, the Deep Neural Networks in the cases where the training data are not abundant. There are several techniques which allow us to prevent the overfitting, e.g., L1/L2 regularization, unsupervis

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::e566d2f4b4fa94fbb1cb93702cec89ae
https://doi.org/10.1007/978-3-319-68456-7_17

Zobrazit plný text záznamu

On Continuous Space Word Representations as Input of LSTM Language Model

Autor: Ludĕk Müller, Daniel Soutner

Publikováno v: Statistical Language and Speech Processing ISBN: 9783319257884
SLSP

Artificial neural networks have become the state-of-the-art in the task of language modelling whereas Long-Short Term Memory LSTM networks seem to be an efficient architecture. The continuous skip-gram and thei¾źcontinuous bag of words CBOW are alg

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::e4f71d7b0794c4a78430f26c1e48a0c7
https://doi.org/10.1007/978-3-319-25789-1_25

Zobrazit plný text záznamu

On a Hybrid NN/HMM Speech Recognition System with a RNN-Based Language Model

Autor: Daniel Soutner, Jan Zelinka, Luděk Müller

Publikováno v: Speech and Computer ISBN: 9783319115801
SPECOM

In this paper, we present a new NN/HMM speech recognition system with a NN-base acoustic model and RNN-based language model. The employed neural-network-based acoustic model computes posteriors for states of context-dependent acoustic units. A recurr

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::f5286a79ec31f3ef252d91170b4258a3
https://doi.org/10.1007/978-3-319-11581-8_39

Zobrazit plný text záznamu

Continuous Distributed Representations of Words as Input of LSTM Network Language Model

Autor: Daniel Soutner, Luděk Müller

Publikováno v: Text, Speech and Dialogue ISBN: 9783319108155
TSD

The continuous skip-gram model is an efficient algorithm for learning quality distributed vector representations that are able to capture a large number of syntactic and semantic word relationships. Artificial neural networks have become the state-of

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::28a7b60e16e57523d0616f291622fd55
https://doi.org/10.1007/978-3-319-10816-2_19

Zobrazit plný text záznamu

Application of LSTM Neural Networks in Language Modelling

Autor: Luděk Müller, Daniel Soutner

Publikováno v: Text, Speech, and Dialogue ISBN: 9783642405846
TSD

Artificial neural networks have become state-of-the-art in the task of language modelling on a small corpora. While feed-forward networks are able to take into account only a fixed context length to predict the next word, recurrent neural networks (R

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::5896593ca8a5db7cd5cd3e578e172908
https://doi.org/10.1007/978-3-642-40585-3_14

Zobrazit plný text záznamu

Neural Network Language Model with Cache

Autor: Luděk Müller, Aleš Pražák, Zdeněk Loose, Daniel Soutner

Publikováno v: Text, Speech and Dialogue ISBN: 9783642327896
TSD

In this paper we investigate whether a combination of statistical, neural network and cache language models can outperform a basic statistical model. These models have been developed, tested and exploited for a Czech spontaneous speech data, which is

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::33e37675894f337b8914665952487669
https://doi.org/10.1007/978-3-642-32790-2_64

Zobrazit plný text záznamu

Web Text Data Mining for Building Large Scale Language Modelling Corpus

Autor: Daniel Soutner, Jan Švec, Jan Hoidekr, Jan Vavruška

Publikováno v: Text, Speech and Dialogue ISBN: 9783642235375
TSD

The paper describes a system for collecting a large text corpus from Internet news servers. The architecture and text preprocessing algorithms are described. We also describe the used duplicity detection algorithm. The resulting corpus contains more

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::a7f89e5ecf8ad4fca6d55ffab46d92f2
https://doi.org/10.1007/978-3-642-23538-2_45

Zobrazit plný text záznamu

Vyhledávací nástroje:

Upřesnit hledání