• The Journal of the Korea institute of electronic communication sciences
• /
• v.5 no.6
• /
• pp.568-574
• /
• 2010

In an effort to improve the speech recognition rate, we investigated performance comparison between speaker-independent and gender-specific speech recognitions. For this purpose, 20 male and 20 female speakers each pronounced 300 isolated Korean words and the speeches were divided into 4 groups: female, male, and two mixed genders. To examine the validity for the gender-specific speech recognition, Fourier spectrum and MFCC feature vectors averaged over male and female speakers separately were examined. The result showed distinction between the two genders, which supports the motivation for the gender-specific speech recognition. In experiments of speech recognition rate, the error rate for the gender-specific case was shown to be less than50% compared to that of the speaker-independent case. From the obtained results, it might be suggested that hierarchical recognition of gender and speech recognition might yield better performance over the current method of speech recognition.

• The Journal of the Acoustical Society of Korea
• /
• v.28 no.4
• /
• pp.383-391
• /
• 2009

In speech signal processing, speech signal corrupted by noise should be enhanced to improve quality. Usually noise estimation methods need flexibility for variable environment. Noise profile is renewed on silence region to avoid effects of speech properties. So we have to preprocess finding voice region before noise estimation. However, if received signal does not have silence region, we cannot apply that method. In this paper, we proposed SNR estimation method for continuous speech signal. The waveform which is stationary region of voiced speech is very correlated by pitch period. So we can estimate the SNR by correlation of near waveform after dividing a frame for each pitch. For unvoiced speech signal, vocal track characteristic is reflected by noise, so we can estimate SNR by using spectral distance between spectrum of received signal and estimated vocal track. Lastly, energy of speech signal is mostly distributed on voiced region, so we can estimate SNR by the ratio of voiced region energy to unvoiced.

• The Journal of the Acoustical Society of Korea
• /
• v.23 no.2
• /
• pp.180-186
• /
• 2004

The presence of noise in speech signals degrades the performance of recognition systems in which there are mismatches between the training and test environments. To make a speech recognizer robust, it is necessary to compensate these mismatches. In this paper, we studied about an improvement of stochastic feature extraction based on band-SNR for robust speech recognition. At first, we proposed a modified version of the multi-band spectral subtraction (MSS) method which adjusts the subtraction level of noise spectrum according to band-SNR. In the proposed method referred as M-MSS, a noise normalization factor was newly introduced to finely control the over-estimation factor depending on the band-SNR. Also, we modified the architecture of the stochastic feature extraction (SFE) method. We could get a better performance when the spectral subtraction was applied in the power spectrum domain than in the mel-scale domain. This method is denoted as M-SFE. Last, we applied the M-MSS method to the modified stochastic feature extraction structure, which is denoted as the MMSS-MSFE method. The proposed methods were evaluated on isolated word recognition under various noise environments. The average error rates of the M-MSS, M-SFE, and MMSS-MSFE methods over the ordinary spectral subtraction (SS) method were reduced by 18.6%, 15.1%, and 33.9%, respectively. From these results, we can conclude that the proposed methods provide good candidates for robust feature extraction in the noisy speech recognition.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1998.06c
• /
• pp.147-150
• /
• 1998

In the case of speech synthesis, the waveform coding method with high quality is mainly used to the synthesis by analysis. Because the parameters of this coding method are not classified as both excitation and vocal tract parameters, it is difficult to apply the waveform coding method to the synthesis by rule. Thus, in order to apply the waveform coding method to the synthesis by rule, a pitch alteration is required for the prosody control. In the speech synthesis method by the conventional PSOLA technique, applying symmetric window function to asymmetric speech waveform, it occurs the unbalance phenomenon of energy according to the overlapped degree of pitch interval adjustment. In this paper to overcome the unbalance phenomenon of energy, we proposed a new method that can convert asymmetric waveform to symmetric one by time-frequency conversion. As a result, we can obtain an average spectrum distortion ratio with 6.38% according to the pitch alteration ratio.

• Journal of the Korean Institute of Telematics and Electronics
• /
• v.25 no.4
• /
• pp.430-436
• /
• 1988

LSF(Line Spectral Frequency) derived from LPC has known as a very useful transmission parameter of speech signals, for it has a good linear interpolation characteristics and a low spectrum distortion at low bit rates coding. This paper presents that it is possible to extract directly the formant frequencies of speech signals from LSF parameter without application of FFT algorithm by comparing the distribution of LSF parameter with the frequency distribution of analysis filter. This paper suggests the advanced algorithm that results in improving the speed of convergence at analytic solution method. Also, for the flexibility of parameters, the process that transforms from LSF to LPC is presented.

• The Journal of the Acoustical Society of Korea
• /
• v.20 no.4E
• /
• pp.24-30
• /
• 2001

This paper describes a fast harmonic estimation, referred to as Delta Adjustment (DA), using a low resolution pitch. The presented DA method is based on modification of the Generalized Dual Excitation (GDE) technique[1] which was proposed to improve speech enhancement performance. We introduce the GDE technique and modify it to be suitable for low bit rate harmonic coding that uses only an integer pitch estimate. Unlike the GDE, the DA matches a frequency-warped version of the original spectrum that conforms to a fixed pitch at all harmonic bands. In addition, complexity and performance of the presented method are described in comparison with those of the conventional Fractional Pitch (FP) based harmonic estimation. Experimental results showed that the DA algorithm significantly reduces the complexity of the FP method while maintaining the performance.

• Proceedings of the IEEK Conference
• /
• 2003.11a
• /
• pp.49-52
• /
• 2003

In this paper, we propose new speech feature parameter using NMf(Non-Negative Matrix Factorization). NMF can represent multi-dimensional data based on effective dimensional reduction through matrix factorization under the non-negativity constraint, and reduced data present parts-based features of input data. In this paper, we verify about usefulness of NMF algorithm for speech feature extraction applying feature parameter that is got using NMF in Mel-scaled filter bank output. According to recognition experiment result, we could confirm that proposal feature parameter is superior in recognition performance than MFCC(mel frequency cepstral coefficient) that is used generally.

• Phonetics and Speech Sciences
• /
• v.6 no.3
• /
• pp.149-153
• /
• 2014

In frequency domain ICA, the frequency bin permutation problem falls off the quality of separated signals. In this paper, we propose a new algorithm to solve the frequency bin permutation problem using the covariance of power ratio of separated signals in multi-channel FD-ICA. It makes use of the continuity of the spectrum of speech signals to check if frequency bin permutation occurs in the separated signal using the power ratio of adjacent frequency bins. Experimental results have shown that the proposed method could fix the frequency bin permutation problem in the multi-channel FD-ICA.

• 제어로봇시스템학회:학술대회논문집
• /
• 1987.10b
• /
• pp.473-478
• /
• 1987

The purpose of this study is to design a phoneme-class recognition system for Korean language using filter-bank analysis and zero crossing rate method. First, the speech signals are separated in 16 bandpass filters to obtain short-time spectrum of speech signals, and digitized by 16-ch A/D converter. And then, with the set of features which extracted from patterns of ratios of each channel energy level to overall energy level, the decision rules are made for recognize unknown speech signal. In this experiment, the recognition rate was about 93.1 percent for 7 vowels under multitalker environment and 74.4 percent for 10 initial sounds at single speaker.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.19 no.4
• /
• pp.761-771
• /
• 1994

This paper focuses on how spectrum information can be represented efficiently in a very low bit rate CELP speech coder. To achieve the goal, an LPC cepstral coefficients VQ scheme representing the spectrum information in a CELP coder is proposed. To represent the spectrum information using LPC cepstrums, three different cepstral distance measures having different spectral meanings in the frequency domain are considered, and their performances are compared and analyzed. The experimental results show that spectrum information in low bit rate CELP coders can be represented very efficiently using the proposed LPC cepstral vector quantization scheme.