Paper Info
Title
The Cambridge University 2014 Bolt Conversational Telephone Mandarin Chinese Lvcsr System For Speech Translation
Abstract
This paper presents the development of the 2014 Cambridge University conversational telephone Mandarin Chinese LVCSR system for the DARPA BOLT speech translation evaluation. A range of advanced modelling techniques were employed to both improve the recognition performance and provide a suitable integration with the translation system. These include an improved system combination technique using frame level acoustic model combination via joint decoding. Sequence trained deep neural network (DNN) based hybrid and tandem systems were combined on-the-fly to produce a consistent decoding output during search. A multi-level paraphrastic recurrent neural network LM (RNNLM) modelling both alternative paraphrase expressions and character sequences while preserving a consistent character to word segmentation was also used. This system gave an overall character error rate (CER) of 29.1% on the BOLT dev14 development set.
Year
Venue
Keywords
2015
16TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION (INTERSPEECH 2015), VOLS 1-5
conversational speech transcription, speech translation, system combination, RNNLM, character LM
Field
DocType
Citations 
Expression (mathematics),Computer science,Word error rate,Recurrent neural network,Text segmentation,Speech recognition,Natural language processing,Artificial intelligence,Speech translation,Artificial neural network,Mandarin Chinese,Acoustic model
Conference
5
PageRank 
References 
Authors
0.42
21
6
Name
Order
Citations
PageRank
Xunying Liu133052.46
Federico Flego2556.19
Linlin Wang3263.91
Chao Zhang4959.70
Mark J. F. Gales53905367.45
Philip C. Woodland64097488.66