Seed funding for strategic research @ RTP (research manpower)
Modern voice authentication systems perform extremely well on large population high quality clean speech databases. While novel algorithms can be designed to provide performance and accuracy, the performance degrades rapidly in the presence of noise. Noise introduces a mismatch between the verificat...
| Yazar: | |
|---|---|
| Diğer Yazarlar: | |
| Materyal Türü: | Research Report |
| Dil: | English |
| Baskı/Yayın Bilgisi: |
2009
|
| Konular: | |
| Online Erişim: | http://hdl.handle.net/10356/17242 |
| _version_ | 1826119647494144000 |
|---|---|
| author | Thambipillai Srikanthan. |
| author2 | School of Computer Engineering |
| author_facet | School of Computer Engineering Thambipillai Srikanthan. |
| author_sort | Thambipillai Srikanthan. |
| collection | NTU |
| description | Modern voice authentication systems perform extremely well on large population high quality clean speech databases. While novel algorithms can be designed to provide performance and accuracy, the performance degrades rapidly in the presence of noise. Noise introduces a mismatch between the verification utterance and the speaker template which causes unpredictable scores leading to performance degradation. Our research attempts to address the problem of mismatch condition caused by additive noise. We have proposed novel algorithms for noise compensation in the speaker model domain and demonstrated their efficiency on TIMIT database corrupted with additive noise. Subsequently, we have combined the proposed algorithm with spectral subtraction method to further improve the performance of the authentication have been successfully translated to dedicated hardware architecture and prototyped on FPGA-based platforms. |
| first_indexed | 2024-10-01T05:03:32Z |
| format | Research Report |
| id | ntu-10356/17242 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T05:03:32Z |
| publishDate | 2009 |
| record_format | dspace |
| spelling | ntu-10356/172422023-03-03T20:22:01Z Seed funding for strategic research @ RTP (research manpower) Thambipillai Srikanthan. School of Computer Engineering DRNTU::Engineering::Computer science and engineering::Computing methodologies::Pattern recognition Modern voice authentication systems perform extremely well on large population high quality clean speech databases. While novel algorithms can be designed to provide performance and accuracy, the performance degrades rapidly in the presence of noise. Noise introduces a mismatch between the verification utterance and the speaker template which causes unpredictable scores leading to performance degradation. Our research attempts to address the problem of mismatch condition caused by additive noise. We have proposed novel algorithms for noise compensation in the speaker model domain and demonstrated their efficiency on TIMIT database corrupted with additive noise. Subsequently, we have combined the proposed algorithm with spectral subtraction method to further improve the performance of the authentication have been successfully translated to dedicated hardware architecture and prototyped on FPGA-based platforms. RGM 11/04 2009-06-02T01:54:16Z 2009-06-02T01:54:16Z 2008 2008 Research Report http://hdl.handle.net/10356/17242 en 39 p. application/pdf |
| spellingShingle | DRNTU::Engineering::Computer science and engineering::Computing methodologies::Pattern recognition Thambipillai Srikanthan. Seed funding for strategic research @ RTP (research manpower) |
| title | Seed funding for strategic research @ RTP (research manpower) |
| title_full | Seed funding for strategic research @ RTP (research manpower) |
| title_fullStr | Seed funding for strategic research @ RTP (research manpower) |
| title_full_unstemmed | Seed funding for strategic research @ RTP (research manpower) |
| title_short | Seed funding for strategic research @ RTP (research manpower) |
| title_sort | seed funding for strategic research rtp research manpower |
| topic | DRNTU::Engineering::Computer science and engineering::Computing methodologies::Pattern recognition |
| url | http://hdl.handle.net/10356/17242 |
| work_keys_str_mv | AT thambipillaisrikanthan seedfundingforstrategicresearchrtpresearchmanpower |