Speaker independent recognition on OLLO French corpus by using different features
2010 (English)In: Pervasive Computing, Signal Processing and Applications (PCSPA 2010), IEEE , 2010, 332-335 p.Conference paper (Refereed)
The Oldenburg LOgatome speech corpus (OLLO) is specifically designed for evaluating speech recognition methods on variability. The performance of features carried on intrinsic variabilities in speech is meaningful for automatic speech recognition (ASR) system. ZCPA and MFCC were the two main features applied to OLLO French corpus in this paper. We took cepstral mean subtraction (CMS) on MFCC. Dynamic transforms (delta-delta-ZCPA and delta-delta-MFCC) were also adopted. The experiments show that the MFCC outperform the ZCPA in separate style. But ZCPA is more robust between different variabilities. The delta-delta operation of MFCC achieves best recognition in noise-free environment. Moreover, ZCPA could be complementary to MFCC so that one can combine them together especially on soft speaking style.
Place, publisher, year, edition, pages
IEEE , 2010. 332-335 p.
MFCC, OLLO corpus, ZCPA, delta-delta-MFCC
IdentifiersURN: urn:nbn:se:liu:diva-91452DOI: 10.1109/PCSPA.2010.87ISBN: 978-1-4244-8043-2ISBN: e-978-0-7695-4180-8OAI: oai:DiVA.org:liu-91452DiVA: diva2:617968
First International Conference on Pervasive Computing, Signal Processing and Applications (PCSPA 2010), Harbin, China, 17-19 September 2010