| EPSRC Reference: |
GR/N01569/01 |
| Title: |
NOVEL FEATURES FOR SPEAKER VERIFICATION |
| Principal Investigator: |
Naylor, Professor PA |
| Other Investigators: |
|
| Researcher Co-Investigators: |
|
| Project Partners: |
|
| Department: |
Electrical and Electronic Engineering |
| Organisation: |
Imperial College London |
| Scheme: |
Standard Research (Pre-FEC) |
| Starts: |
12 June 2000 |
Ends: |
11 June 2003 |
Value (£): |
144,297
|
| 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 |
|
Speaker verification aims to verify the identity of an individual from their voice. Important applications include Internet commerce, telephone banking and access control. We are particularly interested in remote operation.Existing approaches base the classification on spectral features of the speech. Our approach is to model the speech in terms of a voice-source and channel and then base the classification on the parameters of the model. This has the advantage that the voice-source parameters employed are influenced to a lesser degree by phonetic context (giving low intra-speaker variability), exploit the significant variation between speakers' voice-source properties (giving high inter-speaker variability) and are difficult to mimic since the characteristics of the voice-source are not under direct concious muscle control (giving robustness to imposters).The proposed research will develop new algorithms for extracting voice-source features with special focus on robustness to spectral distortions and noise such as encountered on telephone networks. This will include work on closed/open-phase detection in the larynx cycle for which a new energy-based approach is already under development. The relative merits of two alternative methods for combining voice-source features with conventional features will then be investigated. Evaluations will be performed on industry-standard test-sets.
|
|
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.imperial.ac.uk |