Signal and Data Processing

Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:

---

The design for new feature extraction methods out of the speech signal and combination of their obtained information is one of the most effective approaches to improve the performance of automatic speech recognition (ASR) system. Recent researches have been shown that the speech signal contains nonlinear and chaotic properties, but the effects of these properties are not used in the continuous ASR systems. Reconstructed phase space (RPS) is an appropriate domain to exhibit nonlinear properties of a chaotic signal. Therefore, in this paper a new method is proposed to utilize the RPS-based features (LLRPS). These features will be computed using similarity scores between the embedded speech signal in the RPS and a set of predefined phoneme manifolds. Then, TMLP-based neural network estimates phoneme posterior probability over the LLRPS features. The used neural network includes proper properties such as extracting dynamic information and output combination methods. Experimental results using Farsdat speech database show that nonlinear combination of the speech recognition outputs including traditional MFCC features and the LLRPS features, leading to improvement of 3.94% and 4.02% in the accuracy of frame and phoneme recognition, respectively.

Keywords: Continuous speech recognition, Feature extraction, Reconstructed phase space, Phoneme manifolds, Likelihood Score, Neural network

Full-Text [PDF 3564 kb]   (2753 Downloads)    

Add your comments about this article : Your username or Email:

---

---