Kernel eigenvoice
{{Multiple issues|
{{technical|date=July 2018}}
{{context|date=July 2018}}
}}
Speaker adaptation is an important technology to fine-tune either features or speech models for mis-match due to inter-speaker variation. In the last decade, eigenvoice (EV) speaker adaptation has been developed. It makes use of the prior knowledge of training speakers to provide a fast adaptation algorithm (in other words, only a small amount of adaptation data is needed). Inspired by the kernel eigenface idea in face recognition, kernel eigenvoice (KEV) is proposed.{{Cite web |url=http://www.cs.ust.hk/~mak/PG-Thesis/thesis-simon.pdf |title=Kernel Eigenvoice Thesis |access-date=2009-07-17 |archive-url=https://web.archive.org/web/20110610141128/http://www.cs.ust.hk/~mak/PG-Thesis/thesis-simon.pdf |archive-date=2011-06-10 |url-status=dead }} KEV is a non-linear generalization to EV. This incorporates Kernel principal component analysis, a non-linear version of Principal Component Analysis, to capture higher order correlations in order to further explore the speaker space and enhance recognition performance.
See also
References
External links
- [http://en.scientificcommons.org/45651640 Kernel Eigenvoice Speaker Adaptation], ScientificCommons
- {{cite conference |last1=Mak |first1=B. |last2=Ho |first2=S. |date=2005 |title=Various Reference Speakers Determination Methods for Embedded Kernel Eigenvoice Speaker Adaptation |book-title=IEEE International Conference on Acoustics, Speech, and Signal Processing, 2005. Proceedings |conference=ICASSP '05 |volume=1 |pages=981–984 |doi=10.1109/ICASSP.2005.1415280}}
- {{cite journal |last1=Mak |first1=B. |last2=Kwok |first2=J. T. |last3=Ho |first3=S. |date=September 2005 |title=Kernel Eigenvoice Speaker Adaptation |url=http://www.cse.ust.hk/~jamesk/papers/sap05.html |journal=IEEE Transactions on Speech and Audio Processing |volume=13 |issue=5 |pages=984–992 |issn=1063-6676 |doi=10.1109/TSA.2005.851971 |s2cid=7361772 |accessdate=2017-11-15|url-access=subscription }}
- [http://www.cse.ust.hk/~jamesk/papers/icslp04.pdf Speedup of Kernel Eigenvoice Speaker Adaptation by Embedded Kernel PCA], ICSLP 2004.
- [http://papers.nips.cc/paper/2421-eigenvoice-speaker-adaptation-via-composite-kernel-principal-component-analysis.pdf Speaker Adaptation via Composite Kernel PCA], NIPS 2003.
- {{cite journal |last1=Mak |first1=Brian Kan-Wing |last2=Hsiao |first2=Roger Wend-Huu |last3=Ho |first3=Simon Ka-Lung |last4=Kwok |first4=J. T. |date=July 2006 |title=Embedded kernel eigenvoice speaker adaptation and its implication to reference speaker weighting |journal=IEEE Transactions on Audio, Speech, and Language Processing |volume=14 |issue=4 |pages=1267–1280 |doi=10.1109/TSA.2005.860836|citeseerx=10.1.1.206.4596 |s2cid=7527119 }}