Connectionist Speech Recognition: A Hybrid Approach
- H. Bourlard, N. Morgan
- Computer Science
- 1 October 1993
From the Publisher:
Connectionist Speech Recognition: A Hybrid Approach describes the theory and implementation of a method to incorporate neural network approaches into state-of-the-art continuous…
Social signal processing: Survey of an emerging domain
- A. Vinciarelli, M. Pantic, H. Bourlard
- Computer ScienceImage and Vision Computing
- 1 November 2009
Microphone array post-filter based on noise field coherence
- I. McCowan, H. Bourlard
- PhysicsIEEE Transactions on Speech and Audio Processing
- 1 November 2003
A more general expression of the post-filter estimation is developed based on an assumed knowledge of the complex coherence of the noise field that can be used to construct a more appropriate post- filter in a variety of different noise fields.
A mew ASR approach based on independent processing and recombination of partial frequency bands
- H. Bourlard, S. Dupont
- Computer ScienceProceeding of Fourth International Conference on…
- 3 October 1996
The preliminary results presented in this paper show that such an approach, even using quite simple recombination strategies, can yield at least comparable performance on clean speech while providing better robustness in the case of noisy speech.
Text detection, recognition in images and video frames
- Datong Chen, J. Odobez, H. Bourlard
- Computer SciencePattern Recognition
- 1 March 2004
Robust speaker change detection
- J. Ajmera, I. McCowan, H. Bourlard
- PhysicsIEEE Signal Processing Letters
- 26 July 2004
A criterion which can be used to identify speaker changes in an audio stream without such tuning is presented, which consists of calculating the LLR of two models with the same number of parameters.
Towards increasing speech recognition error rates
- H. Bourlard, H. Hermansky, N. Morgan
- Computer ScienceSpeech Communication
- 18 September 1995
On the Use of Information Retrieval Measures for Speech Recognition Evaluation
- I. McCowan, Darren Moore, H. Bourlard
- Computer Science
- 2004
It is suggested that posing speech recognition evaluation as an information retrieval problem, where each word is one unit of information, offers a flexible framework for application-oriented performance analysis based on the concepts of recall and precision.
Modeling human interaction in meetings
- I. McCowan, Samy Bengio, H. Bourlard
- PsychologyIEEE International Conference on Acoustics…
- 6 April 2003
This paper investigates the recognition of group actions in meetings by modeling the joint behaviour of participants using HMM-based approaches and initial results demonstrate the ability of the system to recognise the set of meeting actions.
Continuous speech recognition by connectionist statistical methods
- H. Bourlard, N. Morgan
- Computer ScienceIEEE Trans. Neural Networks
- 1 November 1993
The authors review the basic principles of their hybrid HMM/MLP approach and describe a series of improvements that are analogous to the system modifications instituted for the leading conventional HMM systems over the last few years.
...
...