Báo cáo hóa học: " Research Article Mapping Speech Spectra from Throat Microphone to Close-Speaking Microphone: A Neural Network Approach" potx

... pages doi:10.1155/2007/87219 Research Article Mapping Speech Spectra from Throat Microphone to Close-Speaking Microphone: A Neural Network Approach A. Shahina 1 and B. Yegnanarayana 2 1 Department of Computer Science and Engineering, ... perceptual quality of the throat microphone speech by mapping the speech spectra from the throat microphone t...

Ngày tải lên: 22/06/2014, 19:20

... frame-by-frame basis. If the channel condition changes largely during a frame, the channel estimation is likely to be inaccurate, and the system performance can be severely degraded. To avoid this situation, ... alarm probability is defined as the probability that at least one false alarm is issued during T limit . Let us denote the final false alarm and the final missdetection probabiliti...

Ngày tải lên: 21/06/2014, 05:20

... forth. 4.2. Radio Broadcast Database. For the purpose of our analysis a radio broadcast database was built. The database contains radio broadcast material collected from several public radio stations. ... on radio broadcast database, (b) Var. of SF feature on radio broadcast database, (c) VMFBE feature on BNSI database and (d) Var. of SF feature on BNSI database. the second-rated feature...

Ngày tải lên: 21/06/2014, 19:20

... 2001. [2] Canadian Association of Speech- Language Pathologists and Audiologists, “General Speech & Hearing Fact Sheet,” Report, http://www.caslpa.ca/PDF/fact%20sheets/ speechhearingfactsheet.pdf. [3] ... movement. Vocal quality is the same as of spastic dysarthria. Voice pauses associated with dystonia may occur. Hypernasality is common. This type of dysarthria could lead to a tot...

Ngày tải lên: 21/06/2014, 20:20

... 39-male and 2-female laryngectomees. Their average age was 62.0 ± 7.7 years. They had undergone total laryngectomy because of T3 or T4 laryngeal or hypopharyngeal cancer at least one year prior to ... estimation scales such as visual analogue scales, because the interval size of an equal-appearing inter- val scale might be not equal in all continua. Hence, statistical analyses which do not...

Ngày tải lên: 21/06/2014, 20:20

... covered with a large number of cross-terms apart from the autoterms. In the sequel, the achieved payload and some related applications are given. 4.1. Data payload In this example, we have used a single ... digital audio watermarking,” IEEE Transactions on Multimedia, vol. 7, no. 2, pp. 212–221, 2005. [24] S. Xiang and J. Huang, “Histogram-based audio watermarking against time-scale mod...

Ngày tải lên: 21/06/2014, 22:20

... mapping/ coding codesign. ACKNOWLEDGMENTS This work was supported by NSERC, Canada, (under Alexander Graham Bell Canada Graduate Scholarship and research Grant 249704-07). Part of this work was ... leads to a suboptimal detection and consequently the BICM capacity is always smaller than the CM capacity. M. Benjillali and L. Szczecinski 5 Table 1: Minimum SNR required to attain the sp...

Ngày tải lên: 21/06/2014, 23:20

... are eliminated manually and the remaining shapes are labelled and stored in the database as binary images representing the nine configurations (Figure 18). The training and test sets of the database ... so that phonemes which have similar lip shapes can be differentiated. Thanks to the combination of both lip shapes and manual gestures, each phoneme has a specific visual aspect. Such a “ha...

Ngày tải lên: 22/06/2014, 00:20

... each phoneme has a specific visual aspect. Such a “hand and lip reading” becomes as meaningful as the oral message. The interest of CS is to use a code which is similar to oral language. As a ... The transition shapes are eliminated manually and the remaining shapes are labelled and stored in the database as binary images representing the nine configurations (Figure 18). The trainin...

Ngày tải lên: 22/06/2014, 06:20

... [27]. 3.3. Canonical correlation analysis Canonical correlation analysis (CANCOR) is a multivari- ate statistical analysis allowing to jointly transform the au- dio and visual feature spaces while maximizing ... visual note. However, to our knowledge, ICA has never been used as a transformation of the audiovisual speech feature space (as in [26] for the piano). A Matlab im- plemen...

Ngày tải lên: 22/06/2014, 19:20