A speech enhancement method using adaptive speech PDF

Thanhikam, Weerawut; Kawamura, Arata; Iiguni, Youji

doi:10.1109/ispacs.2011.6146108

Cited by 1 publication

(5 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the speech enhancement system, we used a 50% overlapping frame with 64 samples at 8 kHz sampling frequency. We set a = 0.9 8, ez = 7, yz = 10, which a coefficient is defined in [5], ez and yz are noise estimation parameters defined in [4]. Fig.…”

Section: Resultsmentioning

confidence: 99%

“…In the method we chose for implementation, various histogram are created from speech data from low to high SNR intervals [5]. Arata Kawamura found the optimal value for both parameters that give the best fit of the speech PDF to the speech histogram in each SNR intervals.…”

Section: Mmentioning

confidence: 99%

“…This method gives a better noise suppression during non-speech segment. However, the shape of the speech PDF in a speech segment may be incorrectly estimated, because the shape parameters are determined from only two histograms which are made from speech signals in high and low SNRs, respectively [5].…”

mentioning

confidence: 99%

“…In this paper, algorithm proposed at [5] has been examined and evaluated using MA TLAB and the result shows preeminent characteristics to the conventional algorithms. However, most topics definitely evaluate noise enhancement through simulation, so they can not estimate practical problems, e.g., speed, resource and power consumption, which are important factors in most real digital systems, especially, handheld devices.…”

mentioning

confidence: 99%

“…In this paper, we propose a entirely new hardware architecture implements Adaptive Speech PDF method proposed at [5]. A full system for noise cancellation has been implemented on Xilinx's FPGA for performance evaluation purpose.…”

mentioning

confidence: 99%

See 4 more Smart Citations

A novel hardware architecture for Noise Cancellation based on adaptive speech probability density function

Nguyen

Lam

Hoang

et al. 2014

2014 International Conference on Advanced Technologies for Communications (ATC 2014)

View full text Add to dashboard Cite

A novel hardware architecture for Noise Cancellation (NC) based on adaptive speech probability density function (PDF) is proposed in this paper. The used PDF has two adaptive shape parameters which affect the quality of enhanced speech. Noise can be efficiently suppressed by estimating properly these parameters so that the adaptive speech PDF shape fits to the real speech PDF one. A novel hardware architecture is proposed based on adaptive speech PDF algorithm which shows good results in implementation. Proposed hardware system with stage-pineline technique has achieved a maximum clock frequency of 127 MHz and processing latency is 8.6uS, this enables noise cancellation processing can happen in real-time with very low clock frequency. Hardware scheduling technique is also applied to reduce operators for getting low-resource. A parameter controller configured by system firmware is also implemented to make NC system more flexible. Experimental results show that proposed NC system can suppress noise in real time with high quality of enhanced speech.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Mmentioning

confidence: 99%

mentioning

confidence: 99%

mentioning

confidence: 99%

mentioning

confidence: 99%

See 3 more Smart Citations