| EPSRC Reference: |
GR/M87146/01 |
| Title: |
AN INTEGRATED MULTIPLE-LEVEL STATISTICAL MODEL FOR SPEECH PATTERN PROCESSING |
| Principal Investigator: |
Russell, Professor M |
| Other Investigators: |
|
| Researcher Co-Investigators: |
|
| Project Partners: |
|
| Department: |
Electronic, Electrical and Computer Eng |
| Organisation: |
University of Birmingham |
| Scheme: |
Standard Research (Pre-FEC) |
| Starts: |
07 February 2000 |
Ends: |
06 February 2003 |
Value (£): |
179,084
|
| EPSRC Research Topic Classifications: |
| Human Communication in ICT |
|
|
| EPSRC Industrial Sector Classifications: |
| Creative Industries |
No relevance to Underpinning Sectors |
|
| Related Grants: |
|
| Panel History: |
|
|
Summary on Grant Application Form |
|
The impressive improvements in speech recognition performance which Hidden Markov Model-based systems have achieved on controlled, large vocabulary tasks have not been matched by fundamental improvements in our approach to speech modelling. We still do not understand how to incorporate speech knowledge as a computationally useful constraint in a practical model for speech recognition. The goal of this research is to develop a novel statistical framework for speech pattern modelling, where the relationship between symbolic and acoustic representations of speech is regulated by an articulatory-based intermediate representation which captures inherent constraints of the speech production process. This will require a new, rigorous theory of data-driven, multiple level statistical modelling, and will involve the derivation of a mathematical framework, extension of training and recognition algorithms, and experimental evaluation on a standard speech corpus.If successful, better modelling of phenomena such as co-articulation and articulatory effort should result in improved recognition of natural speech. Models which avoid assumptions of random variation should offer improved noise robustness, and the distillation of speaker characteristics into a compact intermediate representation should facilitate fast speaker-adaptation. Success would also constitute a significant step towards a unified framework for speech pattern modelling, capable of supporting recognition and synthesis.
|
|
Key Findings |
|
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
|
|
Potential use in non-academic contexts |
|
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
|
|
Impacts |
|
| Description |
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk |
| Summary |
|
| Date Materialised |
|
|
|
|
Sectors submitted by the Researcher |
|
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
|
| Project URL: |
|
| Further Information: |
|
| Organisation Website: |
http://www.bham.ac.uk |