Filters
Bottle-Neck Feature Extraction Structures for Multilingual Training and Porting
This article describes the Bottle-Neck feature extraction structures for multilingual training and porting.
Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)
- 2016 •
- D •
- Link
Rok uplatnění
D - Stať ve sborníku
Výsledek na webu
Tensorflow implementation of speaker recognition with x-vector topology
This is a Tensorflow implementation of x-vector topology (speaker embedding). It uses Kaldi toolkit for data processing. We train the model using Tensorflow and also extract speaker embeddings (x-vectors) using it. This allow to train to mod...
Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)
- 2019 •
- R •
- Link
Rok uplatnění
R - Software
Výsledek na webu
Various DNN-HMM architectures used in acoustic modeling with single-speaker and single-channel
recognition tasks. A typical LF-MMI training procedure uses a special 1-state HMM topology topologies (1-, 2- and 3-state HMM topologies that have outputs associated with states of the amount of training data and the conte...
Automation and control systems
- 2021 •
- D •
- Link
Rok uplatnění
D - Stať ve sborníku
Výsledek na webu
Boosting Performance on Low-resource Languages by Standard Corpora: AN ANALYSIS
In this paper, we have evaluated the multilingual techniques for single source-language scenario. Since it is hard to obtain coherent multilingual corpora usable for multilingual training, using single, well resourced, language instead is quite attra...
Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)
- 2016 •
- D •
- Link
Rok uplatnění
D - Stať ve sborníku
Výsledek na webu
Complexity of the TDNN Acoustic Model with Respect to the HMM Topology
recognition tasks. Some of the key features of the LF-MMI approach are: training DNN rate and the use of a simpler HMM topology. The conventional 3-state HMM topology was replaced in a typical LF-MMI training procedure wit...
Automation and control systems
- 2020 •
- D •
- Link
Rok uplatnění
D - Stať ve sborníku
Výsledek na webu
Semi-supervised DNN training with word selection for ASR
The article is about semi-supervised DNN training with word selection for Automatic Speaker Recognition (ASR).
Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)
- 2017 •
- D •
- Link
Rok uplatnění
D - Stať ve sborníku
Výsledek na webu
Voltammetric Determination of Genotoxic Dinitronaphthalenes by Meniscus Modified Silver Solid Amalgam Electrode
Optimum conditions are described for the determination of genotoxic 1,3-, 1,5- and 1,8-dinitronaphthalenes (DNN) by differential pulse voltammetry (DPV were found to be 2.10(-6) Mol l(-1) for 1,3-DNN, 1(.)10(-6) Mol l(-1) for 1,5-
CF - Fyzikální chemie a teoretická chemie
- 2007 •
- Jx
Rok uplatnění
Jx - Nezařazeno - Článek v odborném periodiku (Jimp, Jsc a Jost)
Semi-supervised Training of Deep Neural Networks
Our quest in this paper is to search for an optimal dataselection strategy for the semi-supervised DNN training. We performed an analysis at all the three stages of DNN training....
Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)
- 2013 •
- D •
- Link
Rok uplatnění
D - Stať ve sborníku
Výsledek na webu
TFApprox: Towards a Fast Emulation of DNN Approximate Hardware Accelerators on GPU
Energy efficiency of hardware accelerators of deep neural networks (DNN) can, a software emulator of the DNN accelerator is usually executed on CPU or GPU. However, this emulation is typically two or three orders of magnitude slower...
Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)
- 2020 •
- D •
- Link
Rok uplatnění
D - Stať ve sborníku
Výsledek na webu
Automatic Language Identification Using Deep Neural Networks
In this work, we experimented with the use of deep neural networks (DNNs) to automatic language identification (LID). Guided by the success of DNNs for acoustic modelling, we explored their capability to learn discriminative languag...
Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)
- 2014 •
- D •
- Link
Rok uplatnění
D - Stať ve sborníku
Výsledek na webu
- 1 - 10 out of 5 790