• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2017.06a
• /
• pp.105-106
• /
• 2017

본 논문에서는 최대 우도 기반 파라미터 생성 알고리즘을 적용하여 인공 신경망의 출력인 음향 파라미터 열의 정확성 및 자연성을 향상시키는 방법을 제안하였다. 인공 신경망의 출력으로 정적 특징벡터 뿐 만 아니라 동적 특징벡터도 함께 사용하였고, 미리 계산된 파라미터 분산을 파라미터 생성에 사용하였다. 추정된 정적, 동적 특징벡터의 평균, 분산을 EM 알고리즘에 적용하여 최대 우도 기준 파라미터를 추정할 수 있다. 제안된 알고리즘은 파라미터 생성 시 동적 특징벡터 및 분산을 함께 적용하여 시간축에서의 자연성을 향상시켰다. 제안된 알고리즘의 객관적 평가로 MCD, F0 의 RMSE 를 측정하였고, 주관적평가로 선호도 평가를 실시하였다. 그 결과 기존 알고리즘 대비 객관적, 주관적 성능이 향상되는 것을 검증하였다.

• Proceedings of the Acoustical Society of Korea Conference
• /
• spring
• /
• pp.295-298
• /
• 2000

제품의 음질 특성과 관련된 주요 주파수 대역을 파악하기 위해 본 연구에서는 청소기 소리의 주파수 대역을 음성인식에 중요한 주파수 대역인 4개의 대역으로 나누고, 각 대역 성분을 직교배열표에 따라 가감한 16개의 소리를 만들었다 만들어진 소리에 대해 10명을 대상으로 4가지의 표현어로 Semantic Differential Method(SDM)로 주관적 평가를 하여, 청소기음의 주파수 특성과의 상관관계를 살펴보았다. 불쾌한 느낌과 관련이 깊은 주파수 대역은 고주파수 대역이었으며, 성능이 좋은 느낌을 주기 위해서는 저주파수 대역을 증가시키는 것 이 효과적이었다.

• The Journal of the Acoustical Society of Korea
• /
• v.17 no.8
• /
• pp.82-87
• /
• 1998

음성합성은 합성방식에 따라 파형부호화법, 신호원부호화법, 혼성부호화법으로 분류 할 수 있다. 특히 고음질 합성을 위해서는 파형부호화를 이용한 합성방식이 적합하다. 그렇 지만, 파형부호화를 이용한 합성법은 여기 성분과 여파기 성분을 분리하지 않고 처리하기 때문에 음절단위나 음소단위의 합성기법으로는 바람직하지 못하다. 따라서 파형부호화법을 규칙에 의한 합성에 적용되도록 음원피치를 변경시키기 위한 피치 변경법이 필요하게 된다. 본 논문에서는 스펙트럼 왜곡을 최소화하기 위해 켑스트럼의 성질을 이용하여 피치를 변경 하는 방법에 대하여 제안하였다. 이 방법은 주파수영역상에서 여기 스펙트럼과 여파기 스펙 트럼을 분리하여 여기 스펙트럼을 여기 켑스트럼으로 변환한 후 영값 삽입이나 삭제에 의해 피치를 변경하고 스펙트럼영역에서 피치 변경된 스펙트럼을 재구성하는 기법을 적용하였다. 제안한 방법의 성능을 평가하기 위해 스펙트럼 왜곡율을 측정하여 본 결과 평균 스펙트럼 왜곡율은 평균 2.29%이하로 유지되었으며 주관적인 음질도 평균 3.74로 우수하였다.

• The Journal of the Acoustical Society of Korea
• /
• v.28 no.2
• /
• pp.148-154
• /
• 2009

In audio coding, SBR technology synthesizes the high-bands using patched time-frequency information from low-bands and the correction parameters, Since SBR transmits only correction parameters for high-bands, it provides a low-rate coding of high-bands, and is used as a core module of MPEG-4 HE-AAC, SBR was originally designed for audio signal and its performance for speech signal tends to decrease, and the major reason is an excessive noise floor in high-bands which is caused by incorrect tonality computation, In this paper, a new method to determine noise floor level in an adaptive fashion according to the speech characteristics is proposed in order to solve the problem of SBR for speech signal, The proposed method maintains the compatibility with the standard SBR, and the subjective performance evaluation shows that the proposed method improves the SBR performance especially for male speech signal compared with the standard SBR.

• The Journal of the Acoustical Society of Korea
• /
• v.23 no.5
• /
• pp.400-409
• /
• 2004

In this paper. a voice personality transformation method is proposed. which makes one person's voice sound like another person's voice. To transform the voice personality. vocal tract transfer function is used as a transformation parameter. Comparing with previous methods. the proposed method makes transformed speech closer to target speaker's voice in both subjective and objective points of view. Conversion between vocal tract transfer functions is implemented by classification of entire vector space followed by linear transformation for each cluster. LPC cepstrum is used as a feature parameter. A joint classification and transformation method is proposed, where optimum clusters and transformation matrices are simultaneously estimated in the sense of a minimum mean square error criterion. To evaluate the performance of the proposed method. transformation rules are generated from 150 sentences uttered by three male and on female speakers. These rules are then applied to another 150 sentences uttered by the same speakers. and objective evaluation and subjective listening tests are performed.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.35 no.2
• /
• pp.134-140
• /
• 2015

Traditional moxibustion therapy can cause severe pain and leave scarring burns at the moxibustion site as it relies on the practitioner's subjective and qualitative treatment. Recently, ultrasound therapy has received attention as an alternative to moxibustion therapy owing to its objectiveness and quantitative nature. However, in order to convert ultrasound energy into heat energy, there is a need to precisely understand the ultrasound-focusing characteristics of the acoustic lens. Therefore, in this study, an FEM simulation was performed for acoustic lenses with different geometries a concave lens and zone lens as the geometry critically influences ultrasound focusing. The acoustic pressure field, amplitude, and focal point were also calculated. Furthermore, the performance of the fabricated acoustic lens was verified by a sound pressure measurement experiment.

• The Journal of the Acoustical Society of Korea
• /
• v.20 no.3
• /
• pp.42-47
• /
• 2001

This paper improves the performance of VAD (Voice Activity Detector) in G.723.1 Annex A 6.3kbps/5.3kbps dual rate speech coder, which is developed for Internet Phone and videoconferencing. The VAD decision is based on a three-level energy threshold. We evaluates for processing time, speech quality, and bit rate. The processing time is reduced due to the accuracy of VAD decision on the silence period. On subjective quality test there is almost no difference compared with the G.723.1. In order to measure the bit rate we count the active speech frame (VAD=1) and we can reduce more bit rate as silence periods are shown.

• The Journal of the Acoustical Society of Korea
• /
• v.23 no.6
• /
• pp.481-487
• /
• 2004

In this paper. the structure of bandwidth scalable wideband speech codec and its high-band codec are proposed. In the high-band codec. the signal is divided into frequency bands. and each band is quantized in DCT domain. The DCT coefficients are splitted into magnitude and sign, and each is quantized independently by a specialized method based on its characteristics. In addition. the quantized gain parameter in the low-band codec is utilized in the high-band codec for an enhanced performance. The bandwidth scalable wideband speech codec using G.729E for low-band and the proposed codec for high-band is developed, and it is confirmed that the proposed codec has better subjective performance than 24kbps G.722.1.

• The Journal of the Acoustical Society of Korea
• /
• v.32 no.3
• /
• pp.227-236
• /
• 2013

In this paper, we present a vocal suppression algorithm that can enhance the quality of music signal coded using Spatial Audio Object Coding (SAOC) in Karaoke mode. The residual vocal component in the coded music signal is estimated by using a cross prediction method in which the music signal coded in Karaoke mode is used as the primary input and the vocal signal coded in Solo mode is used as a reference. However, the signals are extracted from the same downmix signal and highly correlated, so that the music signal can be severely damaged by the cross prediction. To prevent this, a psycho-acoustic disturbance rule is proposed, in which the level of disturbance to the reference input of the cross prediction filter is adapted according to the auditory masking property. Objective and subjective test were performed and the results confirm that the proposed algorithm offers improved quality.

• The Journal of the Acoustical Society of Korea
• /
• v.20 no.3
• /
• pp.98-103
• /
• 2001

The binaural auditory system of human has the capability of differentiating the direction and distance of sound sources. This feature is well characterised in terms of the inter-aural intensity difference (IID), the inter-aural time difference (ITD) and/or the spectral shape difference (SSD) arising from the acoustic transfer of a sound source to the outer ears. This paper proposes an effective way of extracting the three sound perception factors (IID, ITD, SSD) from the head-related transfer functions (HRTF's) that depends on the direction and distance of the acoustic source from the listener. It includes the estimation method of the equivalent ITD and 1/3-octave band-based IID factors and their usage to locate a sound source in space. Subjective and objective tests were carried out to examine the effectiveness of the proposed methodology and its applicability to real sound systems. Those experimental results are illustrated in this paper.