| EPSRC Reference: |
GR/L60463/02 |
| Title: |
MULTI-RESOLUTION PHONETIC FEATURE AND MODELS FOR SPONTANEOUS SPEECH RECOGNITION |
| Principal Investigator: |
Vaseghi, Professor S |
| Other Investigators: |
|
| Researcher Co-Investigators: |
|
| Project Partners: |
|
| Department: |
Information Systems & Computing |
| Organisation: |
Brunel University London |
| Scheme: |
Standard Research (Pre-FEC) |
| Starts: |
01 January 2000 |
Ends: |
30 September 2000 |
Value (£): |
33,664
|
| EPSRC Research Topic Classifications: |
| Human Communication in ICT |
|
|
| EPSRC Industrial Sector Classifications: |
| No relevance to Underpinning Sectors |
|
|
| Related Grants: |
|
| Panel History: |
|
|
Summary on Grant Application Form |
|
The objective of this research proposal is to develop robust phonetic features and models for improving the accuracy of automatic speech recognition. The new features and models should enable hidden Markov models to better utilise the acoustic cues and the varying signal to noise ratios across the spectral-time trajectory of phonetic units of speech. The initial foci of our research are the follow:(1) development of a set of multi-resolution cepstral features extracted at several different levels of spectral resolutions,(2) development of discriminative phonetic features for the utilisation of the correlations along the trajectory of phonetic units of speech, and(3) development of modified hidden Markov models, and the application of decision-tree clustering method, for the modelling of the effects of increased coarticulation in spontaneous speech. The features will be integrated and optimised using a minimum classification error training process. The application focus of the programme is the recognition of phonetic units in spontaneous speech where the recognition accuracy is generally lower than that of the continuous (relatively well articulated) speech. The research will in particular investigate the influence of a higher level of coarticulation in spontaneous speech on the distribution of phonetic features and models.
|
|
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.brunel.ac.uk |