Informatics Report Series
|
|
|
|
|
|
Title:Connectionist Speech Recognition of Broadcast News |
Authors:
A J Robinson
; G D Cook
; D P W Ellis
; J E Fosler-Lussier
; Steve Renals
; D A G Williams
|
Date: 2002 |
Publication Title:Speech Communication |
Publisher:Elsevier |
Publication Type:Journal Article
Publication Status:Published
|
Volume No:# 37(1-2)
Page Nos:27-45
|
DOI:10.1016/S0167-6393(01)00058-9
|
- Abstract:
-
This paper describes connectionist techniques for recognition of Broadcast News. The fundamental difference between connectionist systems and more conventional mixture-of-Gaussian systems is that connectionist models directly estimate posterior probabilities as opposed to likelihoods. Access to posterior probabilities has enabled us to develop a number of novel approaches to confidence estimation,
pronunciation modelling and search. In addition we have investigated a new feature extraction technique
based on the modulation-filtered spectrogram, and methods for combining multiple information sources. We have incorporated all of these techniques into a system for the transcription of Broadcast News, and we present results on the 1998 DARPA Hub-4E Broadcast News evaluation data.
- Links To Paper
- 1st Link
- Bibtex format
- @Article{EDI-INF-RR-0662,
- author = {
A J Robinson
and G D Cook
and D P W Ellis
and J E Fosler-Lussier
and Steve Renals
and D A G Williams
},
- title = {Connectionist Speech Recognition of Broadcast News},
- journal = {Speech Communication},
- publisher = {Elsevier},
- year = 2002,
- volume = {# 37(1-2)},
- pages = {27-45},
- doi = {10.1016/S0167-6393(01)00058-9},
- url = {http://www.cstr.ed.ac.uk/downloads/publications/2002/specom02-preprint.pdf},
- }
|