• 대한의용생체공학회:의공학회지
• /
• 제43권2호
• /
• pp.102-108
• /
• 2022

Purpose: To propose a deep learning-based clinical decision support technique for laryngeal disease on epiglottis, tongue and vocal cords. Materials and Methods: A total of 873 laryngeal endoscopic images were acquired from the PACS database of Pusan N ational University Yangsan Hospital. and VGG16 model was applied with transfer learning and fine-tuning. Results: The values of precision, recall, accuracy and F1-score for test dataset were 0.94, 0.97, 0.95 and 0.95 for epiglottis images, 0.91, 1.00, 0.95 and 0.95 for tongue images, and 0.90, 0.64, 0.73 and 0.75 for vocal cord images, respectively. Conclusion: Experimental results demonstrated that the proposed model have a potential as a tool for decision-supporting of otolaryngologist during manual inspection of laryngeal endoscopic images.

• 대한후두음성언어의학회지
• /
• 제19권1호
• /
• pp.11-15
• /
• 2008

Endoscopic laser cordectomy is known as an oncologically sound procedure for T1 and selected T2 glottic carcinoma ; it has comparable local control rate and better long-term laryngeal preservation rate when compared with those of radiotherapy. Even if results of the reported voice outcome studies after surgery or radiotherapy are diverse and controversial, resection deeper than the body layer of the vocal fold (type III, IV, V cordectomy) usually leads to aerodynamic insufficiency during phonation and results in poor voice quality. A keyhole defect or development of synechiae at the anterior commissure after type VI cordecomy may also result in unsatisfactory vocal outcome. However, many advances in phonosurgical techniques are reported to be successfully applied in the reconstruction of glottal defect that is subsequent to endoscopic laser cordectomy. In case of glottal insufficiency, voice restoration can be achieved by means of augmentation of the paraglottic space or medialization of the excavated vocal fold. Injection laryngoplasty with synthetic materials or autologous fat is gaining its popularity for restoring minor glottal volume defect because of its convenience. Laryngeal framework surgery, especially type I thyroplasty with premade implant systems or Gore-Tex, is most frequently used to correct larger glottic volume defect. In case of anterior commissural keyhole defect, additional procedure including laryngofissure may be required. For anterior commissural synechiae, laryngeal keel may be inserted for several weeks or mitomycin-C may be repeatedly applied after the division of adhesive scar to prevent restenosis. In this paper, current concepts and the authors' experiences of phonosurgical reconstruction of vocal function after endoscopic cordectomy will be introduced.

• 한국지능시스템학회논문지
• /
• 제19권6호
• /
• pp.773-778
• /
• 2009

본 논문에서는 인간의 감정 변화에 강인한 음성 인식 시스템을 구현하기 위하여 감정 변화의 영향을 최소화 하는 방법에 관한 연구를 수행하였다. 이를 위하여 우선 다양한 감정이 포함된 음성 데이터베이스를 사용하여 감정 변화에 따른 음성 신호의 변화를 관찰하였다. 감정이 포함되지 않은 평상의 음성으로 학습된 음성 인식 시스템에 감정이 포함된 인식 데이터가 입력되는 경우 감정에 따른 음성의 차이가 인식 시스템의 성능을 저하시킨다. 본 연구에서는 감정의 변화에 따라 화자의 성도 길이가 변화한다는 것과 이러한 변화는 음성 인식 시스템의 성능을 저하시키는 원인 중의 하나임을 관찰하였다. 본 연구에서는 이러한 음성의 변화를 감소시키는 방법으로 성도 길이 정규화 방법을 사용한 감정 변화에 강인한 음성 인식 시스템을 개발하였다. HMM을 사용한 단독음 인식 실험에서 제안된 학습 방법을 사용하면 감정 데이터의 오차가 기존방법보다 41.9% 감소되었다.

• 대한후두음성언어의학회지
• /
• 제30권1호
• /
• pp.15-20
• /
• 2019

Background and Objectives : The goal of this study is to present a strategy for improving the self-regulation (SR) ability and facilitating the change of vocal behavior by applying voice therapy using the SR concept to the patients with vocal cord nodule and muscle tension dysphonia. Materials and Method : The subjects were 80 patients and 80 patients who were diagnosed with muscle tension dysphonia and vocal nodules. As a control group, the results were compared among patients with the same dysphonia without using SR strategies. The concept of SR before voice therapy was explained to the patients, and the treatment was divided into three stages according to the goal of voice therapy. The treatment stages consist of 1) skill acquisition, 2) habit formation, and 3) habit changes. voice therapy was performed by applying SR strategies such as goal implementation intentions and a less routine behavior. Patient's dropout rates were measured to compare the adherence of voice therapy. Results : Significant improvement was seen in all groups receiving voice therapy. However, in the group using the SR strategy, the voice analysis results showed a relatively low dropout rate of voice therapy. In the generalization confirmation stage, patients who applied SR concept showed better results. SR strategy did no longer be necessary to maintain newly adopted vocal behavior. Conclusion : The results of this study show that SR is one of the cognitive factors that can have a significant impact on the outcome of voice therapy, and also has a positive impact on the acquisition and generalization of new skills. A better understanding of SR and the development of therapeutic strategies using it will play an important role in solving voice problems in clinical settings.

• 한국환경생태학회지
• /
• 제24권2호
• /
• pp.202-208
• /
• 2010

본 연구는 박쥐의 연령 증가에 따른 발성 초음파의 변화를 파악하기 위하여 집박쥐(Pipistrellus abramus)를 대상으로 생후 어린 개체가 발산하는 음을 이용하여 세부 분석을 실시하였다. 분석결과 어린 개체는 성체와 다른 복합적이고 다양한 형태의 시그널을 발산하는 것으로 확인되었다. 펄스의 형태는 생후 2일차에는 하모니의 수가 많고 불규칙적인 시그널 형태를 보였으며, 생후 10일차부터 20일 까지는 완만한 FM 시그널 형태를 보였다. 생후 30일 경에는 짧은 시간의 FM 시그널 형태를 보였으며, 40일 이후 점차 성체와 유사한 형태의 시그널을 발산하였다. 어린개체의 성장에 따른 발성 시그널을 분석한 결과 각각의 분석 항목별로 유의적인 차이가 있는 것으로 나타났다. 즉, 펄스의 지속시간(p<0.001), 펄스 간격(p<0.001), 하모니의 수(p<0.001)는 새끼의 연령이 증가할수록 감소하는 경향을 보였으며, 최고진동수(p<0.001), 개시진동수(p<0.001), 종말진동수(p<0.001)는 증가하는 결과를 보였다. 특히 펄스 간격은 생후 5일차에 가장 큰 변화를 보였으며, 그 외 펄스 지속시간, 최고진동수, 개시진동수, 종말진동수는 생후 25일차에 가장 큰 변화를 나타내었다.

• 제어로봇시스템학회논문지
• /
• 제16권12호
• /
• pp.1137-1142
• /
• 2010

This paper studied the speech parameters less affected by the human emotion for the development of the robust speech recognition system. For this purpose, the effect of emotion on the speech recognition system and robust speech parameters of speech recognition system were studied using speech database containing various emotions. In this study, mel-cepstral coefficient, delta-cepstral coefficient, RASTA mel-cepstral coefficient and frequency warped mel-cepstral coefficient were used as feature parameters. And CMS (Cepstral Mean Subtraction) method were used as a signal bias removal technique. Experimental results showed that the HMM based speaker independent word recognizer using vocal tract length normalized mel-cepstral coefficient, its derivatives and CMS as a signal bias removal showed the best performance of 0.78% word error rate. This corresponds to about a 50% word error reduction as compare to the performance of baseline system using mel-cepstral coefficient, its derivatives and CMS.

• 대한후두음성언어의학회지
• /
• 제26권1호
• /
• pp.21-24
• /
• 2015

Voice therapy after post-thyroidectomy dysphonia is designed to improve glottal closure without supraglottic hyperfunction by development of abdominal support for breathing and intrinsic muscle strengthening exercises. Regarding voice therapy of unilateral vocal fold paresis/paralysis, several techniques such as head positioning, digital manipulation, pushing and pulling method are used for trial or temporary therapy. And the holistic voice therapy such as LSVT, Accent method, VFE are useful for hypofunctional voice disorders. These may be effective interventions as a temporary improvement until normal voice recovers or may provide the essentials for long-term rehabilitation.

• 대한후두음성언어의학회지
• /
• 제15권1호
• /
• pp.10-15
• /
• 2004

Background and Objective : Voice disorders are one of the major occupational hazards of school teaching. Prolonged voice use through verbal instruction has been implicated as a cause of vocal impairment among members of this profession. They all depend on their voice for their livelihood and are greatly alarmed by slightest alteration in voice quality. Despite considerable research in the area of voice problems in teachers, the prevalence of voice disorders in this group is unknown. Severe voice problems can affect teacher's ability to teach in the classroom. Materials and Method : This study investigated the prevalence of self-reported voice problems in teachers using a survey of a simple random sample of middle school teachers (n=1000). As part of the survey, teachers were asked to report voice problems for the day of the survey, duration, main symptom, smoking, caffeine, alcohol history etc. Results : The reponse rate was 91%(n=94) with 89.7%(n=812) teachers reporting voice problems on the day of the survey, and 40%(n=330) of these group reporting the history of medical consult. Conclusion : These finding indicate a need for vocal hygiene education in teachers and for the development of educational programs aimed at preventing voice problems in this group of professional voice users.

• 한국지능시스템학회논문지
• /
• 제22권6호
• /
• pp.681-686
• /
• 2012

본 논문에서는 강인한 감정 음성 인식 시스템을 개발하기 위하여 감정의 영향을 적게 받는 음성 파라메터에 대한 연구를 수행하였다. 이러한 목적을 위하여 다양한 감정이 포함된 데이터를 사용하여 감정이 음성 인식 시스템과 음성 파라메터에 미치는 영향을 분석하였다. 본 연구에서는 멜 켑스트럼, 델타 멜 켑스트럼, RASTA 멜 켑스트럼, 루트 켑스트럼, PLP 계수와 성도 길이 정규화 방법에서 주파수 와핑된 멜 켑스트럼 계수를 사용하였다. 또한 신호 편의 제거 방법으로 CMS 방법과 SBR 방법이 사용되었다. 실험결과에서 성도정규화 방법을 사용한 RASTA 멜 켑스트럼, 델타 멜 켑스트럼 및 CMS 방법을 사용한 경우가 HMM 기반의 화자독립 단독음 인식 실험 결과에서 가장 우수한 결과를 나타내었다.

• International journal of thyroidology
• /
• 제11권2호
• /
• pp.109-116
• /
• 2018

갑상선 수술 후 발생한 성대마비는 삶의 질을 중요시하는 요즘 시대에 제일 치명적 합병증 중 하나이다. 그 결과 수술 중 신경감시시스템의 사용은 보편화되고 있으나 아직 기존의 시스템은 외과 의사가 사용하기에 불편하거나 문제점들이 있다. 그래서 새로운 방식의 신경 탐침과 신경 감시 장치의 개발이 필요한 시점이다. 이에 최근 모든 수술 기구(금속형 기구, 내시경 및 로봇 기구, 에너지 기반 디바이스)에 탈부착이 가능한 신경을 자극하는 신경 탐침 및 후두 떨림을 측정하기 위한 표면압력센서를 이용한 새로운 형태의 수술 중 신경감시시스템의 개발에 대한 연구가 기대된다.