Limits...
An Undecimated Wavelet-based Method for Cochlear Implant Speech Processing.

Hajiaghababa F, Kermani S, Marateb HR - J Med Signals Sens (2014)

Bottom Line: The undecimated wavelet packet transform (UWPT) is computed like the wavelet packet transform except that it does not down-sample the output at each level.The statistical analysis revealed that the UWT-based N-of-M strategy significantly improved the MOS, STOI and segmental SNR (P < 0.001) compared with what obtained with the IIR filter-bank based strategies.Thus, the information loss is minimal and that is why the UWPT performance was better than that of traditional filter-bank strategies in speech recognition tests.

View Article: PubMed Central - PubMed

Affiliation: Department of Electrical Engineering, Najafabad Branch, Islamic Azad University, Najafabad, Iran.

ABSTRACT
A cochlear implant is an implanted electronic device used to provide a sensation of hearing to a person who is hard of hearing. The cochlear implant is often referred to as a bionic ear. This paper presents an undecimated wavelet-based speech coding strategy for cochlear implants, which gives a novel speech processing strategy. The undecimated wavelet packet transform (UWPT) is computed like the wavelet packet transform except that it does not down-sample the output at each level. The speech data used for the current study consists of 30 consonants, sampled at 16 kbps. The performance of our proposed UWPT method was compared to that of infinite impulse response (IIR) filter in terms of mean opinion score (MOS), short-time objective intelligibility (STOI) measure and segmental signal-to-noise ratio (SNR). Undecimated wavelet had better segmental SNR in about 96% of the input speech data. The MOS of the proposed method was twice in comparison with that of the IIR filter-bank. The statistical analysis revealed that the UWT-based N-of-M strategy significantly improved the MOS, STOI and segmental SNR (P < 0.001) compared with what obtained with the IIR filter-bank based strategies. The advantage of UWPT is that it is shift-invariant which gives a dense approximation to continuous wavelet transform. Thus, the information loss is minimal and that is why the UWPT performance was better than that of traditional filter-bank strategies in speech recognition tests. Results showed that the UWPT could be a promising method for speech coding in cochlear implants, although its computational complexity is higher than that of traditional filter-banks.

No MeSH data available.


Related in: MedlinePlus

Comparison of mean opinion score for undecimated wavelet, and infinite impulse response filter-bank, both with N-of-M, implementations
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4236803&req=5

Figure 2: Comparison of mean opinion score for undecimated wavelet, and infinite impulse response filter-bank, both with N-of-M, implementations

Mentions: Note that in the UWPT, the coarsest resolution is also iteratively decomposed like the fine resolution. The undecimated wavelet approach can be used to decompose the input speech signal into a number of frequency bands. Similar to the FFT-based N-of-M strategy, the number of maximum amplitude channel output, can be selected using a logarithmic compression map and stimulation. A second-order Butterworth low-pass filter (cut-off frequency 400 Hz) was used to obtain smooth envelopes of speech signals. The block diagram of the undecimated wavelet-based N-of-M strategy is shown in Figure 1. In this strategy, input speech signals are passed through a 6-stage wavelet packet decomposition yielding a 64-band output. A channel output is computed by summing up all the frequency-band output falling within the frequency range of the channel and is passed through a rectifier and then low-pass filtered to extract the channel envelope. The number of channels can be varied. The block diagrams of the traditional base CIS and N-of-M strategies were shown in Figures 1 and 2 of the manuscript written by Gopalakrishna et al.[10] Comparing with our proposed N-of-M structure, the FFT block was replaced with the undecimated wavelet and the rectifier and LPF was taken from the CIS strategy.[3]


An Undecimated Wavelet-based Method for Cochlear Implant Speech Processing.

Hajiaghababa F, Kermani S, Marateb HR - J Med Signals Sens (2014)

Comparison of mean opinion score for undecimated wavelet, and infinite impulse response filter-bank, both with N-of-M, implementations
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4236803&req=5

Figure 2: Comparison of mean opinion score for undecimated wavelet, and infinite impulse response filter-bank, both with N-of-M, implementations
Mentions: Note that in the UWPT, the coarsest resolution is also iteratively decomposed like the fine resolution. The undecimated wavelet approach can be used to decompose the input speech signal into a number of frequency bands. Similar to the FFT-based N-of-M strategy, the number of maximum amplitude channel output, can be selected using a logarithmic compression map and stimulation. A second-order Butterworth low-pass filter (cut-off frequency 400 Hz) was used to obtain smooth envelopes of speech signals. The block diagram of the undecimated wavelet-based N-of-M strategy is shown in Figure 1. In this strategy, input speech signals are passed through a 6-stage wavelet packet decomposition yielding a 64-band output. A channel output is computed by summing up all the frequency-band output falling within the frequency range of the channel and is passed through a rectifier and then low-pass filtered to extract the channel envelope. The number of channels can be varied. The block diagrams of the traditional base CIS and N-of-M strategies were shown in Figures 1 and 2 of the manuscript written by Gopalakrishna et al.[10] Comparing with our proposed N-of-M structure, the FFT block was replaced with the undecimated wavelet and the rectifier and LPF was taken from the CIS strategy.[3]

Bottom Line: The undecimated wavelet packet transform (UWPT) is computed like the wavelet packet transform except that it does not down-sample the output at each level.The statistical analysis revealed that the UWT-based N-of-M strategy significantly improved the MOS, STOI and segmental SNR (P < 0.001) compared with what obtained with the IIR filter-bank based strategies.Thus, the information loss is minimal and that is why the UWPT performance was better than that of traditional filter-bank strategies in speech recognition tests.

View Article: PubMed Central - PubMed

Affiliation: Department of Electrical Engineering, Najafabad Branch, Islamic Azad University, Najafabad, Iran.

ABSTRACT
A cochlear implant is an implanted electronic device used to provide a sensation of hearing to a person who is hard of hearing. The cochlear implant is often referred to as a bionic ear. This paper presents an undecimated wavelet-based speech coding strategy for cochlear implants, which gives a novel speech processing strategy. The undecimated wavelet packet transform (UWPT) is computed like the wavelet packet transform except that it does not down-sample the output at each level. The speech data used for the current study consists of 30 consonants, sampled at 16 kbps. The performance of our proposed UWPT method was compared to that of infinite impulse response (IIR) filter in terms of mean opinion score (MOS), short-time objective intelligibility (STOI) measure and segmental signal-to-noise ratio (SNR). Undecimated wavelet had better segmental SNR in about 96% of the input speech data. The MOS of the proposed method was twice in comparison with that of the IIR filter-bank. The statistical analysis revealed that the UWT-based N-of-M strategy significantly improved the MOS, STOI and segmental SNR (P < 0.001) compared with what obtained with the IIR filter-bank based strategies. The advantage of UWPT is that it is shift-invariant which gives a dense approximation to continuous wavelet transform. Thus, the information loss is minimal and that is why the UWPT performance was better than that of traditional filter-bank strategies in speech recognition tests. Results showed that the UWPT could be a promising method for speech coding in cochlear implants, although its computational complexity is higher than that of traditional filter-banks.

No MeSH data available.


Related in: MedlinePlus