• 음성과학
• /
• 제14권2호
• /
• pp.65-71
• /
• 2007

The severe-profound hearing impaired children have various disorders in everday communication due to the lack of hearing feedback. Especially, their speech produced unstable voice, omission and distortion of articulation, pitch break, cul-de-sac voice, and so on so that they were difficult to accurately deliver an intended message. This study attempts to analyze the acoustic characteristics of 4 vowel sounds produced by 35 severe-profound hearing impaired children using CSL(Computerized Speech Lab, Model 4300b). The formant data were obtained from the spectrogram and analyzed data by 12 formant filter and auto-correlation among the formants. Results showed that the hearing impaired children's formant values came out very high. They produced the vowels at the mode of hypertension with unstable voice. In order to improve their speech, they would need some adequate auditory feedback.

• 말소리와 음성과학
• /
• 제7권2호
• /
• pp.55-62
• /
• 2015

It is widely accepted that Parkinson's disease(PD) is the most common cause of hypokinetic dysarthria, and its characteristics of 'short rushes of speech' have become more evident along with the severity of motor disorders. Speech alternate motion rates (AMRs) are particularly useful for observing not only rate abnormalities but also deviant speech. However, relatively little is known about the characteristics of 'short rushes of speech' in terms of AMRs of PD except for the perceptual characteristics. The purpose of this study was to examine which acoustic features of 'short rushes of speech' in terms of AMRs are a robust indicator of Parkinsonian speech. Numbers of syllabic repetitions (/pə/, /tə/, /kə/) in AMR tasks were analyzed through acoustic methods observing a spectrogram of the Computerized Speech Lab in 9 patients with PD. Acoustically, we found three characteristics of 'short rushes of speech': 1) Vocalized consonants without closure duration(VC) 76.3%; 2) No consonant segmentation(NC) 18.6%; 3) No vowel formant frequency(NV) 5.1%. Based on these results, 'short rushes of speech' may affect the failure to reach and maintain the phonatory targets. In order to best achieve the therapeutic goals, and to make the treatment most efficacious, it is important to incorporate training methods which are based on both phonation and articulation.

• 음성과학
• /
• 제11권3호
• /
• pp.131-142
• /
• 2004

Speaker identification system basically functions by comparing spectral energy of an individual production model with that of an input signal. This study aimed to develop a new speaker identification system from two parameters from the spectral energy of numeric sounds: difference sum and correlation coefficient. A narrow-band spectrogram yielded more stable spectral energy across time than a wide-band one. In this paper, we collected empirical data from four male speakers and tested the speaker identification system. The subjects produced 18 combinations of three-digit numeric. sounds !en times each. Five productions of each three-digit number were statistically averaged to make a model for each speaker. Then, the remaining five productions were tested on the system. Results showed that when the threshold for the absolute difference sum was set to 1200, all the speakers could not pass the system while everybody could pass if set to 2800. The minimum correlation coefficient to allow all to pass was 0.82 while the coefficient of 0.95 rejected all. Thus, both threshold levels can be adjusted to the need of speaker identification system, which is desirable for further study.

• 전기학회논문지
• /
• 제60권12호
• /
• pp.2374-2379
• /
• 2011

The fundamental frequency of the voiced speech is estimated using the instantaneous frequency based on the symmetric higher order differential energy operator. The instantaneous frequency based on the symmetric higher order energy operator shows better frequency estimation result since it is aligned to the time instance of the signal. The speech is pre-processed by a lowpass filter to remove higher frequency components. Then, it is processed by the instantaneous frequency to obtain the fundamental frequency estimates. The symmetric higher order energy operator has been used as an indicator to determine the voiced/unvoiced speech. The fundamental frequency estimates are further processed by a moving average filter to obtain the monotonically changed estimates. The obtained fundamental frequency estimates have been compared with the spectrogram of the speech to confirm its accuracy.

• 대한음성학회:학술대회논문집
• /
• 대한음성학회 2006년도 춘계 학술대회 발표논문집
• /
• pp.109-112
• /
• 2006

This study investigated the acoustical characteristics of phonatory offset-onset mechanisms. And this study shows the comparative results between non-stutterers (N=3) and a stutterer (N=1). Phonatory offset-onset means a laryngeal articulatory in the connected speech. In the phonetic context V_V), pattern 0(there is no changes) appeared in all subjects, and pattern 4(this indicate the trace of glottal fry and closure in spectrogram)was only in a Stutterer. In high vowels(/i/, /u/), pattern 3 and 4 appeared only in a stutterer. Although there is no common pattern among the non-stutterers, individual's preference pattern was founded. This study offers the key to an understanding of physiological movement on a block of stutter.

• 대한후두음성언어의학회지
• /
• 제3권1호
• /
• pp.6-12
• /
• 1989

Methods of voice rehabilitation in laryngectomees include training of esophageal speech, use of electrolarynx and pneumatic speech aid and surgical methods, etc. In this paper, we introduce the experimental model of obturator-type electrolarynx which has several advantages for use such as ease of learning, no disagreeable appearance, and both hands not being occupied. We compared it to normal voice and other voice rehabilitation methods such as esophageal voice, japanese pneumatic speech aid and cervical electrolarynx in intelligibility and sonagraphic evaluation. The results are as follows; 1) Obturator-type electrolarynx exhibited the lowest intelligibility. 2) In sonagraphic evaluation, the spectrogram produced by the obturator-type electrolarynx was the most different from those of normal voice.

• 한국음향학회지
• /
• 제40권3호
• /
• pp.234-240
• /
• 2021

본 논문에서는 잡음 환경에서 효과적인 음성 인식을 위해 마스크 기반의 음성 향상 기법을 개선한다. 마스크 기반의 음성 향상 기법에서는 심층 신경망을 기반으로 추정한 마스크를 잡음 오염 음성에 곱하여 향상된 음성을 얻는다. 마스크 추정 모델로 VoiceFilter(VF) 모델을 사용하고 추정된 마스크로 얻은 음성으로부터 잔여 잡음을 보다 확실히 제거하기 위해 Spectrogram Inpainting(SI)기법을 적용한다. 본 논문에서는 음성 향상 결과를 보다 개선하기 위해 마스크 추정을 위한 모델 학습 과정에 사용되는 조합된 손실함수를 제안한다. 음성 구간에 남아 있는 잡음을 보다 효과적으로 제거하기 위해 잡음 오염 음성에 마스크를 적용한 Triplet 손실함수의 Positive 부분을 컴포넌트 손실함수와 조합하여 사용한다. 실험 평가를 위한 잡음 음성 데이터는 TIMIT 데이터베이스와 NOISEX92, 배경음악 잡음을 다양한 Signal to Noise Ratio(SNR) 조건으로 합성하여 만들어 사용한다. 음성 향상의 성능 평가는 Source to Distortion Ratio(SDR), Perceptual Evaluation of Speech Quality(PESQ), Short-Time Objective Intelligibility(STOI)를 이용한다. 실험을 통해 평균 제곱 오차로만 훈련된 기존 시스템과 비교하여, VF 모델은 평균 제곱 오차로 훈련하고 SI 모델은 조합된 손실함수를 사용하였을 때 SDR은 평균 0.5dB, PESQ는 평균 0.06, STOI는 평균 0.002만큼 성능이 향상된 것을 확인했다.

• 문화기술의 융합
• /
• 제7권1호
• /
• pp.640-645
• /
• 2021

딥러닝 기반 종단간 TTS 시스템은 텍스트에서 스펙트로그램을 생성하는 Text2Mel 과정과 스펙트로그램에서 음성신호를 합성하는 보코더 등 두 가지 과정으로 구성되어 있다. 최근 TTS 시스템에 딥러닝 기술을 적용함에 따라 합성음의 명료도와 자연성이 사람의 발성과 유사할 정도로 향상되고 있다. 그러나 기존의 방식과 비교하여 음성을 합성하는 추론 속도가 매우 느리다는 단점을 갖고 있다. 최근 제안되고 있는 비-자기회귀 방식은 이전에 생성된 샘플에 의존하지 않고 병렬로 음성 샘플을 생성할 수 있어 음성 합성 처리 속도를 개선할 수 있다. 본 논문에서는 비-자기회귀 방식을 적용한 Text2Mel 기술인 FastSpeech, FastSpeech 2, FastPitch와, 보코더 기술인 Parallel WaveGAN, Multi-band MelGAN, WaveGlow를 소개하고, 이를 구현하여 실시간 처리 여부를 검증한다. 실험 결과 구한 RTF로 부터 제시된 방식 모두 실시간 처리가 충분히 가능함을 알 수 있다. 그리고 WaveGlow를 제외하고 학습 모델 크기가 수십에서 수백 MB 정도로, 메모리가 제한되어 있는 임베디드 환경에 적용 가능함을 알 수 있다.

• 한국음향학회지
• /
• 제25권8호
• /
• pp.383-388
• /
• 2006

본 논문에서는 잡음 환경에서 단일 마이크로폰의 음성 향상에 대한 새로운 기법을 제시했다. 일반적으로 널리 알려진 스펙트럼 차감법에 근거한 음성 향상 기술은 신호 대 잡음비에 따른 스펙트럼 이득으로 표현된다. 대표적인 Ephraim과 Malah의 decision-directed (DD) 추정치는 잡음 구간에서 효율적으로 뮤지컬 잡음을 제거하지만 음성 구간에서는 이전 프레임의 음성 스펙트럼 성분에 더 큰 비중을 두기 때문에 a priori SNR의 프레임 지연이 발생한다. 따라서 DD에 의해 추정된 a priori SNR이 적용된 잡음 제거 이득은 현재 프레임보다 이전 프레임에 영향을 받으므로 음성 전이 구간에서 잡음 제거 성능을 저하시킨다. 본 논문은 DD의 가중치 파라미터에 Sigmoid Type의 함수를 적용하여 계산적으로는 간단하지만 효과적인 음성 향상 알고리즘을 제안한다. 제안된 접근 방식은 DD의 주요 파라미터인 a priori SNR 지연의 문제점을 해결하면서 뮤지컬 잡음 제거에 우수한 DD의 이점은 유지한다. 제안된 알고리즘의 성능은 다양한 잡음 환경에서 ITU-T P.862 Perceptual Evaluation of Speech Quality (PESQ) 와 Mean Opinion Score (MOS). 그리고 음성 스펙트로그램 (Spectrogram)에 의해 평가했고 기존의 DD의 고정된 가중치 파라미터를 사용했을 때 보다 향상된 결과를 나타내었다.

• 음성과학
• /
• 제12권4호
• /
• pp.53-70
• /
• 2005

A Vowel of speech signals are multicomponent signals composed of AM-FM components whose instantaneous frequency and instantaneous amplitude are time-varying. The changes of emotion states cause the variation of the instantaneous frequencies and the instantaneous amplitudes of AM-FM components. Therefore, it is important to estimate exactly the instantaneous frequencies and the instantaneous amplitudes of AM-FM components for the extraction of key information representing emotion states and changes in speech signals. In tills paper, firstly a method decomposing speech signals into AM - FM components is addressed. Secondly, the fundamental frequency of vowel sound is estimated by the simple method based on the spectrogram. The estimate of the fundamental frequency is used for decomposing speech signals into AM-FM components. Thirdly, an estimation method is suggested for separation of the instantaneous frequencies and the instantaneous amplitudes of the decomposed AM - FM components, based on Hilbert transform and the demodulation property of the extended Fourier transform. The estimates of the instantaneous frequencies and the instantaneous amplitudes can be used for modification of the spectral distribution and smooth connection of two words in the speech synthesis systems based on a corpus.