• MALSORI
• /
• no.66
• /
• pp.61-72
• /
• 2008

This paper proposes a method to improve pathological and normal voice classification performance by combining multiple features such as auditory-based and higher-order features. Their performances are measured by Gaussian mixture models (GMMs) and linear discriminant analysis (LDA). The combination of multiple features proposed by the frame-based LDA method is shown to be an effective method for pathological and normal voice classification, with a 87.0% classification rate. This is a noticeable improvement of 17.72% compared to the MFCC-based GMM algorithm in terms of error reduction.

• The Journal of the Acoustical Society of Korea
• /
• v.14 no.5
• /
• pp.94-101
• /
• 1995

In this paper, a new vocal cords model, called a four-mass model, is proposed for a hoarse speech mechanism. Pathological changes of vocal cords cause hoarse speech and glottal waveform reflects motion states of vocal cords. From these facts, we assumed that the morbid vocal cords be dissymmetric and take the four-mass type. The glottal waveforms and the model parameters of normal and hoarse speech signals are analyzed, and some relations bet ween the model parameters and the hoarse pathology are discussed. Experimental results show that the new research method of hoarse speech can reveal relations between the acoustic features of hoarse speech and the hoarse pathology, and be used to diagnose laryngeal diseases and to improve tone quality of hoarse speech.

• Proceedings of the IEEK Conference
• /
• 2007.07a
• /
• pp.407-408
• /
• 2007

Robust pitch estimation is an important study in many areas of speech processing. In voice pathology, diverse statistics extracted form pitch were commonly used to test voice quality. In this study, we compared several established pitch detection algorithms (PDAs) for verification of adequacy of the PDAs. In the database of total pathological voices of 99 and normal voices of 30, an analysis of errors related with pitch detection was evaluated between pathological and normal voices, or among the types of pathological voices such as benign vocal fold lesions; polyp, nodule, and cysts. Consequently, it is required to survey the severity of tested voice in order to obtain accurate pitch estimates.

• Phonetics and Speech Sciences
• /
• v.7 no.2
• /
• pp.9-17
• /
• 2015

In the present study, we investigated ability of recognition of auditory perception with regards to the quality of voice in postlingual CI adults and proposed a training program to improve within subject reliability. A prospective case-control study was conducted in adults with 7 postlingual deaf who received a CI surgery and 10 normal hearing controls. The pre and post test and training program included parameters of consensus auditory-perceptual evaluation of voice(CAPE-V) with pathological voice sample by using Alvin. In results of pre-post test for monitoring improvements of internal reliability for listeners via the training program, there was statistically significant difference in both test and group. There was statistically significant difference in internal reliability between pre-post test in the normal hearing group, the result was no significant in the CI group. The present study found that CI adults showed less ability in awareness of voice quality compared to normal hearing group. Also the training program improved pitch and loudness in CI adults.

• Speech Sciences
• /
• v.7 no.3
• /
• pp.285-294
• /
• 2000

In this paper the acoustic characteristics of pathological voice, which are measured before and after surgical operation, are compared. This experiment is conducted for the purpose of predicting patients' speech after operation. The voices are recorded from the same patients. Jitter, shimmer and other parameters are. computed and their statistical characteristics are compared. Also spectral changes, such as formant frequency shift and spectral slope change, are compared. From the experimental results, it is verified that not only source characteristics but also vocal tract components vary. And this indicates that the modification of source parameters are not enough for the prediction. Also the result indicates that the operation causes change to both the physical shape of vocal folds and the manner of articulation.

• Speech Sciences
• /
• v.15 no.2
• /
• pp.89-97
• /
• 2008

There is a number of clinical conditions that affect directly or indirectly the physical properties of the vocal folds and thereby the pressure waveforms of elicited sounds. If the relationships between the clinical conditions and the voice quality are sufficiently reliable, it should be possible to detect these diseases or disorders. The focus of this paper is to determine the set of features and their values that would characterize the speaker's state of vocal folds. To the extent that these features can capture the anatomical, physiological, and neurological aspects of the speaker they can be potentially used to mediate an unobtrusive approach to diagnosis. We will show a new approach to this problem supported with results obtained from two disordered voice corpora.

• Speech Sciences
• /
• v.9 no.4
• /
• pp.27-38
• /
• 2002

The purpose of this article is to present the acoustic parameters (VOT, jitter, shimmer, vF0, vAm, NHR, SPI, VTI, DVB, DSH) for consonants (/pipi/, /$p^{h}ip^{h}i$/, /p'ip'i/) and sustained vowels (/a/, /e/, /i/) produced by normal subjects and dysphonia patients at two vocal effort(normal, high) by Lombard effect using 60dB white noise. Lombard effect indicates the vocal effort increase in noisy situation. At normal vocal effort, in general the acoustic parameter values of patients are greater than normal. And in noisy situation, significant decrease of acoustic values is seen in normal compared with in dysphonia patients. The clinical implication of this finding, the vocal quality in dysphonia is not compensated by vocal effort as well as normal subjects because of the inefficiency caused by abnormal vocal fold appearance and function. And with this result, we can counsel that the voice quality can not be improved as well as the patient expect.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1998.06e
• /
• pp.395-398
• /
• 1998

본 논문에서는 웨이브렛 변환에서 구해진 파라미터와 신경회로망을 이용하여 후두의 양성종양과 정상상태를 구분하는 실험을 행하였다. 식별 파라미터로는 웨이브렛변환으로부터 도출된 ECS 파라미터와 jitter, shimmer를 이용하였으며 신경회로망은 한 개의 은닉층을 갖는 다층구조 신경망을 이용하였다. 신경망의 입력으로는 세가지 파라미터의 조합을 두 개 또는 세 개를 입력하여 각각의 경우의 식별율을 조사하였다. 실험결과 75%에서 93%에 이르는 식별율을 얻었다.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1998.08a
• /
• pp.388-391
• /
• 1998

jitter, shimmer 및 켑스트럼 방식의 음원분석에 의한 파라미터를 이용하여 장애음성을 진단, 식별하는 방법을 제안한다. 먼저 통계적 처리결과르 바탕으로 식별에 유효한 파라미터들을 선택하고 이들 파라미터들을 이용하여 최종 진단한다. 식별방법으로는 신경회로망을 이용한다. 입력파라미터로는 jitter, shimmer, HNRR을 사용한다. 신경회로망은 1 은닉층을 갖는 3- layer 신경회로망을 사용한다. 실험결과 효과적으로 정상음성과 장애음성의구분이 가능해졌다.

• Speech Sciences
• /
• v.7 no.1
• /
• pp.77-87
• /
• 2000

Distortion Product Otoacoustic Emissions (DPOAEs) can be measured in the external ear canal two fold: amplitude and latency, but most DPOAE studies deal with amplitude aspects. The purpose of this study was to investigate the latency of the 2f1-f2 DPOAEs in ears with hearing losses and to see if it could be a clinically useful method to distinguish normal from abnormal ears. For this purpose, DPOAE latency were measured as a function of frequency from 1 to 8 kHz in 30 ears with conductive and sensorineural hearing losses (SNHLs). DPOAEs were recorded with Otodynamic Analyzer ILO92. Results showed that the latency decreased as the frequency increased up to 8 kHz. The mean values of DPOAE latency for ears of SNHLs were shorter at all frequencies when they were compared to the mean values of normal ears. The latency in ears of conductive hearing losses was shorter than normal ears at the selective frequencies, as well. The results support the hypothesis that latency values are shorter in pathological ears.