Výsledky vyhledávání - "Dhananjay Ram"

Phonetic subspace features for improved query by example spoken term detection

Autor: Afsaneh Asaei, Hervé Bourlard, Dhananjay Ram

Publikováno v: Speech Communication. 103:27-36

This paper addresses the problem of detecting speech utterances from a large audio archive using a simple spoken query, hence referring to this problem as “Query by Example Spoken Term Detection” (QbE-STD). This still open pattern matching proble

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::6a0a120ed8fd57f11ab7b8cac9bc3e7b
https://doi.org/10.1016/j.specom.2018.07.001

Zobrazit plný text záznamu

Multilingual Bottleneck Features for Query by Example Spoken Term Detection

Autor: Hervé Bourlard, Lesly Miculicich, Dhananjay Ram

Publikováno v: ASRU

State of the art solutions to query by example spoken term detection (QbE-STD) rely on bottleneck feature representation of the query and audio document. Here, we present a study on QbE-STD performance using several monolingual as well as multilingua

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::b991e03f2a18a567218625c499768c4f
https://doi.org/10.1109/asru46091.2019.9003752

Zobrazit plný text záznamu

Document-Level Neural Machine Translation with Hierarchical Attention Networks

Autor: Lesly Miculicich, Nikolaos Pappas, James Henderson, Dhananjay Ram

Publikováno v: EMNLP
arXiv.org e-Print Archive

Neural Machine Translation (NMT) can be improved by including document-level contextual information. For this purpose, we propose a hierarchical attention model to capture the context in a structured and dynamic manner. The model is integrated in the

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::0cf4a72a8b075f48d808611967c35176
https://doi.org/10.18653/v1/d18-1325

Zobrazit plný text záznamu

Self-Attentive Residual Decoder for Neural Machine Translation

Autor: Nikolaos Pappas, Lesly Miculicich Werlen, Andrei Popescu-Belis, Dhananjay Ram

Publikováno v: NAACL-HLT
arXiv.org e-Print Archive
Scopus-Elsevier

Neural sequence-to-sequence networks with attention have achieved remarkable performance for machine translation. One of the reasons for their effectiveness is their ability to capture relevant source-side contextual information at each time-step pre

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::b666210507f60044dd3b10984e58d672

Zobrazit plný text záznamu

Subspace Detection of DNN Posterior Probabilities via Sparse Representation for Query by Example Spoken Term Detection

Autor: Afsaneh Asaei, Hervé Bourlard, Dhananjay Ram

Publikováno v: INTERSPEECH

We cast the query by example spoken term detection (QbE-STD) problem as subspace detection where query and background subspaces are modeled as union of low-dimensional subspaces. The speech exemplars used for subspace modeling are class-conditional p

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::33d62b3126285eadd3fbac7b377ddd4e
https://doi.org/10.21437/interspeech.2016-1278

Zobrazit plný text záznamu

Sparse modeling of posterior exemplars for keyword detection

Autor: Dhananjay Ram, Afsaneh Asaei, Pranay Dighe, Hervé Bourlard

Publikováno v: INTERSPEECH

Sparse representation has been shown to be a powerful modeling framework for classification and detection tasks. In this paper, we propose a new keyword detection algorithm based on sparse representation of the posterior exemplars. The posterior exem

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::e2cb1e67747a0f34d9bb06cdfca9d1f3
https://doi.org/10.21437/interspeech.2015-732

Zobrazit plný text záznamu

A Bayesian approach to speaker normalization using vowel formant frequency

Autor: Rajesh M. Hegde, Dhananjay Ram, Debasis Kundu

Publikováno v: NCC

Large variation in speakers causes significant performance degradation of a speaker independent speech recognition system. In an attempt to compensate for this degradation in performance, this paper proposes a novel Bayesian approach to estimate spea

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::cb3b2170a34f04dbde3a0665cf932cb0
https://doi.org/10.1109/ncc.2014.6811382

Zobrazit plný text záznamu

CNN based Query by Example Spoken Term Detection

Autor: Lesly Miculicich, Dhananjay Ram, Hervé Bourlard

Publikováno v: INTERSPEECH

In this work, we address the problem of query by example spoken term detection (QbE-STD) in zero-resource scenario. State of the art solutions usually rely on dynamic time warping (DTW) based template matching. In contrast, we propose here to tackle

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::14b221bc01d3f2758c4cf931391ff068
https://infoscience.epfl.ch/record/256312

Zobrazit plný text záznamu

Neural Network Based End-to-End Query by Example Spoken Term Detection

Autor: Lesly Miculicich, Hervé Bourlard, Dhananjay Ram

This paper focuses on the problem of query by example spoken term detection (QbE-STD) in zero-resource scenario. State-of-the-art approaches primarily rely on dynamic time warping (DTW) based template matching techniques using phone posterior or bott

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::23726a00f6ee7a165007c6806864eae5
https://infoscience.epfl.ch/record/278156

Zobrazit plný text záznamu

Phonological Posterior Hashing for Query by Example Spoken Term Detection

Autor: Dhananjay Ram, Afsaneh Asaei, Hervé Bourlard

Publikováno v: INTERSPEECH

State of the art query by example spoken term detection (QbE-STD) systems in zero-resource conditions rely on representation of speech in terms of sequences of class-conditional posterior probabilities estimated by deep neural network (DNN). The post

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::005df051b0aef4afbfc8d4ea78fa5e57
https://infoscience.epfl.ch/record/256261

Zobrazit plný text záznamu

Vyhledávací nástroje:

Upřesnit hledání