1 resultado para media networks
em Cambridge University Engineering Department Publications Database
Filtro por publicador
- JISC Information Environment Repository (1)
- Repository Napier (3)
- Aberdeen University (2)
- Abertay Research Collections - Abertay University’s repository (1)
- Academic Archive On-line (Stockholm University; Sweden) (1)
- AMS Tesi di Dottorato - Alm@DL - Università di Bologna (4)
- AMS Tesi di Laurea - Alm@DL - Università di Bologna (1)
- Applied Math and Science Education Repository - Washington - USA (3)
- ArchiMeD - Elektronische Publikationen der Universität Mainz - Alemanha (1)
- Archivo Digital para la Docencia y la Investigación - Repositorio Institucional de la Universidad del País Vasco (1)
- Aston University Research Archive (8)
- B-Digital - Universidade Fernando Pessoa - Portugal (1)
- Biblioteca Digital de la Universidad Católica Argentina (1)
- Biblioteca Digital de Teses e Dissertações Eletrônicas da UERJ (1)
- BORIS: Bern Open Repository and Information System - Berna - Suiça (5)
- Boston University Digital Common (3)
- Cambridge University Engineering Department Publications Database (1)
- CentAUR: Central Archive University of Reading - UK (5)
- Central European University - Research Support Scheme (1)
- Chinese Academy of Sciences Institutional Repositories Grid Portal (1)
- Cochin University of Science & Technology (CUSAT), India (1)
- CORA - Cork Open Research Archive - University College Cork - Ireland (1)
- Deakin Research Online - Australia (30)
- DI-fusion - The institutional repository of Université Libre de Bruxelles (1)
- Digital Commons @ DU | University of Denver Research (1)
- Digital Commons at Florida International University (3)
- Digital Peer Publishing (8)
- DigitalCommons@The Texas Medical Center (1)
- DigitalCommons@University of Nebraska - Lincoln (1)
- DRUM (Digital Repository at the University of Maryland) (1)
- Ecology and Society (1)
- Glasgow Theses Service (1)
- Helda - Digital Repository of University of Helsinki (1)
- Indian Institute of Science - Bangalore - Índia (4)
- Instituto Politécnico do Porto, Portugal (3)
- Memoria Académica - FaHCE, UNLP - Argentina (1)
- Ministerio de Cultura, Spain (1)
- Nottingham eTheses (2)
- Open University Netherlands (1)
- Portal de Revistas Científicas Complutenses - Espanha (8)
- QUB Research Portal - Research Directory and Institutional Repository for Queen's University Belfast (11)
- Queensland University of Technology - ePrints Archive (782)
- RCAAP - Repositório Científico de Acesso Aberto de Portugal (1)
- Repositório Científico do Instituto Politécnico de Lisboa - Portugal (2)
- Repositório digital da Fundação Getúlio Vargas - FGV (5)
- Repositório Institucional da Universidade de Aveiro - Portugal (1)
- Repositório Institucional da Universidade de Brasília (1)
- Repositório Institucional UNESP - Universidade Estadual Paulista "Julio de Mesquita Filho" (1)
- RUN (Repositório da Universidade Nova de Lisboa) - FCT (Faculdade de Cienecias e Technologia), Universidade Nova de Lisboa (UNL), Portugal (1)
- Universidad de Alicante (3)
- Universidad del Rosario, Colombia (3)
- Universidad Politécnica de Madrid (12)
- Universidade Federal do Rio Grande do Norte (UFRN) (1)
- University of Canberra Research Repository - Australia (1)
- University of Southampton, United Kingdom (2)
- University of Washington (1)
- WestminsterResearch - UK (3)
Resumo:
We describe our work on developing a speech recognition system for multi-genre media archives. The high diversity of the data makes this a challenging recognition task, which may benefit from systems trained on a combination of in-domain and out-of-domain data. Working with tandem HMMs, we present Multi-level Adaptive Networks (MLAN), a novel technique for incorporating information from out-of-domain posterior features using deep neural networks. We show that it provides a substantial reduction in WER over other systems, with relative WER reductions of 15% over a PLP baseline, 9% over in-domain tandem features and 8% over the best out-of-domain tandem features. © 2012 IEEE.