|
Department of
Engineering |
|
|
AGILE: Autonomous Global Integrated Language Exploitation
For Postdoctoral positions you will need to complete an
application for
employment cover sheet (PD18) as well as supplying an up-to-date CV.
[
Overview |
Personnel |
Partners |
Publications |
Project Overview and Goals
AGILE is a large, multiple institution project funded by the DARPA
Global Autonomous Language Exploitation (GALE) program with the aim of
developing speech and language processing technologies
to recognise, analyse, and translate speech and text into
readable English.
Research areas include:
- improving the performance of conventional speech recognition systems;
- extracting acoustic and linguistic meta-data from acoustic data to aid
recognition and translation;
- improving the performance of statistical machine translation;
- developing strategies that integrate speech recognition and machine translation.
The research will extend and integrate the current state-of-the-art Speech-to-Text (STT)
and Statistical Machine Translation (SMT) systems developed within
the Speech Research Group at Cambridge University. The
speech recognition systems are based on the Hidden Markov Model Toolkit
(HTK). HTK is a portable toolkit for building and manipulating
hidden Markov models and is in use at hundreds of sites worldwide.
The statistical machine translation work is based on the translation
template model (TTM) system.
top
Project Personnel
Academic Staff:
Prof Phil Woodland [Principal Investigator]
Dr. Bill Byrne [Co-Investigator]
Dr. Mark Gales [Co-Investigator]
Dr. Adrià de Gispert [SMT]
Research Associates:
Dr. Frank Diehl [Arabic STT]
Dr. Andrew Liu [STT and SMT integration]
Dr. Marcus Tomalin [Meta-data, Language Modelling]
Dr. Junho Park [Neural Network Features/Language Modelling]
Research Students:
Graeme Blackwood [Statistical Machine Translation]
Jamie Brunning [Statistical Machine Translation]
Tao Li [OOV detection, pronunciation generation]
CK Raut [Context Aware Speech Processing]
Support Staff:
Anna Langley [Computer Officer]
Former Members:
David Mrva [Continuous Language Models]
Dr. Hichem Sahbi [Statistical Machine Translation]
Dr. Khe Chai Sim [Structured Covariance Modelling]
Dr. Rohit Sinha [STT]
Dr. Lan Wang [Discriminative Adaptive Training]
Dr. Kai Yu [Mandarin STT]
top
Project Partners
Here is a partial list of the AGILE project participants, in addition to the University of Cambridge :
top
Publications
Here is a list of the publications associated with, and funded by, the AGILE project.
[ Speech-to-Text |
Statistical Machine Translation |
STT/SMT Integration ]
Speech-to-Text (STT) - Automatic Speech Recognition
- J. Park, F. Diehl, M. J. F. Gales, M. Tomalin and P. C. Woodland (2009)
Efficient Generation and Use of MLP Features for Arabic Speech Recognition.
INTERSPEECH 2009
- J. Park, F. Diehl, M. J. F. Gales, M. Tomalin and P. C. Woodland (2009)
Training and adapting MLP features for Arabic speech recognition.
ICASSP 2009
- C. K. Raut and M. J. F. Gales (2009)
Bayesian Discriminative Adaptation for
Speech Recognition.
ICASSP 2009
- C. K. Raut, K. Yu and M. J. F. Gales (2008)
Adaptive Training using Discriminative Mapping Transforms.
INTERSPEECH 2008
- M. J. F. Gales, F. Diehl, C. K. Raut, M. Tomalin, P. C. Woodland and K. Yu. (2007)
Development of a phonetic system for large vocabulary Arabic
speech recognition.
ASRU 2007
- K. Yu, M. J. F. Gales and P. C. Woodland. (2007)
Unsupervised training with directed manual transcription for
recognizing Mandarin broadcast audio.
INTERSPEECH 2007
- K. Yu and M.J.F. Gales (2007).
Bayesian Adaptive Inference and Adaptive Training.
IEEE Transactions on Audio, Speech and Language Processing, vol.15, no.6, 1932--1943, 2007.
- X. Liu and M. J. F. Gales (2007).
Automatic Model Complexity Control Using Marginalized Discriminative Growth Functions.
IEEE Transactions on Audio, Speech and Language Processing.
- K.C. Sim and M.J.F. Gales (2006).
Minimum Phone Error Training of Precision Matrix Models.
IEEE Transactions on Speech and Audio Processing, May 2006.
- L. Wang, M.J.F. Gales and P.C. Woodland (2007).
Unsupervised Training for Mandarin Broadcast News and Conversation Transcription.
ICASSP 2007.
- K. Yu and M. J. F. Gales (2006).
Incremental Adaptation Using Bayesian Inference.
IEEE International Conference on Acoustics Speech and Signal Processing 2006.
- Sinha R., Gales M.J.F., Kim D.Y., Liu X.A., Sim K.C. and Woodland P.C.(2006).
The CU-HTK Mandarin Broadcast News Transcription System.
IEEE International Conference on Acoustics Speech and Signal Processing 2006.
- Gales M.J.F., Kim D.Y., Woodland P.C., Chan H.Y. Mrva D., Sinha, R. and Tranter S.E..(2006).
Progress in the CU-HTK Broadcast News Transcription System.
IEEE Transactions Audio Speech and Language Processing, September 2006.
Statistical Machine Translation (SMT)
STT/SMT Integration
- X. Liu, W. Byrne, M. J. F. Gales, A. de Gispert, M. Tomalin, P. C. Woodland and K. Yu. (2007)
Discriminative language model adaptation for Mandarin broadcast
speech transcription and translation.
ASRU 2007
- M.J.F. Gales, X. Liu , R. Sinha, P.C. Woodland, K. Yu, S. Matsoukas T. Ng, K. Nguyen,
L. Nguyen J-L. Gauvain L. Lamel and A. Messaoudi (2007).
Speech Recognition System Combination for Machine Translation.
ICASSP 2007.
top