1 resultado para Smoothing
em Bucknell University Digital Commons - Pensilvania - USA
Filtro por publicador
- Aberystwyth University Repository - Reino Unido (1)
- Academic Archive On-line (Mid Sweden University; Sweden) (1)
- Academic Research Repository at Institute of Developing Economies (2)
- Acceda, el repositorio institucional de la Universidad de Las Palmas de Gran Canaria. España (13)
- AMS Campus - Alm@DL - Università di Bologna (1)
- AMS Tesi di Dottorato - Alm@DL - Università di Bologna (4)
- AMS Tesi di Laurea - Alm@DL - Università di Bologna (3)
- Aquatic Commons (3)
- ArchiMeD - Elektronische Publikationen der Universität Mainz - Alemanha (5)
- Archimer: Archive de l'Institut francais de recherche pour l'exploitation de la mer (1)
- Archive of European Integration (3)
- Archivo Digital para la Docencia y la Investigación - Repositorio Institucional de la Universidad del País Vasco (5)
- Aston University Research Archive (19)
- Biblioteca de Teses e Dissertações da USP (1)
- Biblioteca Digital da Produção Intelectual da Universidade de São Paulo (4)
- Biblioteca Digital da Produção Intelectual da Universidade de São Paulo (BDPI/USP) (4)
- Biblioteca Digital de Teses e Dissertações Eletrônicas da UERJ (2)
- BORIS: Bern Open Repository and Information System - Berna - Suiça (16)
- Boston University Digital Common (2)
- Bucknell University Digital Commons - Pensilvania - USA (1)
- Bulgarian Digital Mathematics Library at IMI-BAS (3)
- CaltechTHESIS (4)
- Cambridge University Engineering Department Publications Database (28)
- CentAUR: Central Archive University of Reading - UK (34)
- Chinese Academy of Sciences Institutional Repositories Grid Portal (32)
- Cochin University of Science & Technology (CUSAT), India (2)
- Collection Of Biostatistics Research Archive (17)
- CORA - Cork Open Research Archive - University College Cork - Ireland (2)
- Corvinus Research Archive - The institutional repository for the Corvinus University of Budapest (1)
- CUNY Academic Works (1)
- Dalarna University College Electronic Archive (1)
- Deakin Research Online - Australia (23)
- Digital Commons - Michigan Tech (1)
- Digital Commons at Florida International University (1)
- Digital Peer Publishing (1)
- DigitalCommons - The University of Maine Research (1)
- DigitalCommons@The Texas Medical Center (1)
- DRUM (Digital Repository at the University of Maryland) (2)
- Duke University (1)
- eResearch Archive - Queensland Department of Agriculture; Fisheries and Forestry (3)
- eScholarship Repository - University of California (1)
- Greenwich Academic Literature Archive - UK (4)
- Helda - Digital Repository of University of Helsinki (6)
- Indian Institute of Science - Bangalore - Índia (36)
- Instituto Politécnico de Castelo Branco - Portugal (1)
- Instituto Politécnico do Porto, Portugal (6)
- Lume - Repositório Digital da Universidade Federal do Rio Grande do Sul (1)
- Massachusetts Institute of Technology (4)
- Memorial University Research Repository (1)
- National Center for Biotechnology Information - NCBI (2)
- Nottingham eTheses (1)
- Portal de Revistas Científicas Complutenses - Espanha (1)
- Publishing Network for Geoscientific & Environmental Data (11)
- QUB Research Portal - Research Directory and Institutional Repository for Queen's University Belfast (28)
- Queensland University of Technology - ePrints Archive (56)
- RDBU - Repositório Digital da Biblioteca da Unisinos (2)
- Repositório Alice (Acesso Livre à Informação Científica da Embrapa / Repository Open Access to Scientific Information from Embrapa) (1)
- Repositório Científico da Universidade de Évora - Portugal (2)
- Repositório Científico do Instituto Politécnico de Lisboa - Portugal (3)
- Repositório digital da Fundação Getúlio Vargas - FGV (17)
- Repositório Institucional da Universidade de Brasília (2)
- Repositório Institucional UNESP - Universidade Estadual Paulista "Julio de Mesquita Filho" (35)
- RUN (Repositório da Universidade Nova de Lisboa) - FCT (Faculdade de Cienecias e Technologia), Universidade Nova de Lisboa (UNL), Portugal (2)
- SAPIENTIA - Universidade do Algarve - Portugal (1)
- Universidad de Alicante (3)
- Universidad del Rosario, Colombia (12)
- Universidad Politécnica de Madrid (28)
- Universidade Complutense de Madrid (1)
- Universidade de Lisboa - Repositório Aberto (1)
- Universidade Federal do Pará (4)
- Universidade Federal do Rio Grande do Norte (UFRN) (8)
- Universitat de Girona, Spain (2)
- Universitätsbibliothek Kassel, Universität Kassel, Germany (3)
- Université de Lausanne, Switzerland (1)
- Université de Montréal (1)
- Université de Montréal, Canada (15)
- University of Canberra Research Repository - Australia (1)
- University of Connecticut - USA (1)
- University of Michigan (2)
- University of Queensland eSpace - Australia (7)
- University of Washington (3)
- Worcester Research and Publications - Worcester Research and Publications - UK (1)
Resumo:
We present a new approach for corpus-based speech enhancement that significantly improves over a method published by Xiao and Nickel in 2010. Corpus-based enhancement systems do not merely filter an incoming noisy signal, but resynthesize its speech content via an inventory of pre-recorded clean signals. The goal of the procedure is to perceptually improve the sound of speech signals in background noise. The proposed new method modifies Xiao's method in four significant ways. Firstly, it employs a Gaussian mixture model (GMM) instead of a vector quantizer in the phoneme recognition front-end. Secondly, the state decoding of the recognition stage is supported with an uncertainty modeling technique. With the GMM and the uncertainty modeling it is possible to eliminate the need for noise dependent system training. Thirdly, the post-processing of the original method via sinusoidal modeling is replaced with a powerful cepstral smoothing operation. And lastly, due to the improvements of these modifications, it is possible to extend the operational bandwidth of the procedure from 4 kHz to 8 kHz. The performance of the proposed method was evaluated across different noise types and different signal-to-noise ratios. The new method was able to significantly outperform traditional methods, including the one by Xiao and Nickel, in terms of PESQ scores and other objective quality measures. Results of subjective CMOS tests over a smaller set of test samples support our claims.