인간의 목소리는 사람간의 정보 전달을 위한 가장 쉬운 방법 중 하나이다. 음성의 특징은 사람마다 다를 수 있으며 발성 속도, 발성기관의 형태와 기능, 피치 톤, 언어 습관 및 성별에 따라 다르게 나타난다. 목소리는 사람의 의사소통 핵심 요소이다. 제 4 차 산업 혁명의 시대에 목소리는 사람과 사람, 사람과 기계, 기계 와 기계 사이의 주요한 의사소통 수단이 된다. 그 이유 때문에 사람들은 자신의 의도를 다른 사람들에게 명확하게 전달하려고 노력한다. 그리고 이 과정에서 목소리는 언어 정보와 함께 다양한 추가 정보가 포함되게 된다. 예를 들어 감정 상태, 건강 상태, 신뢰도와 관련되거나, 거짓말의 여부, 음주로 인한 목소리의 변화 등 다양한 언어 및 비언어 정보를 포함하며, 다양한 분석 파라미터로 나타나게 된다. 이를 활용하면 개인의 신용도를 평가하는 척도로 사용할 수 있다. 특히 성대의 기본 주파수의 특성과 성도의 공진 주파수 특성의 관계를 분석함으로써 얻을 수 있다. 이전의 연구에서 다양한 신용 상태의 변화에 따른 목소리 분석 및 특성 변화를 연구 하였다. 본 연구에서는 음성을 통해 추출 된 매개 변수를 통해 기계 학습을 통한 개인 신용 판별 기를 제안한다.
인간의 목소리는 사람간의 정보 전달을 위한 가장 쉬운 방법 중 하나이다. 음성의 특징은 사람마다 다를 수 있으며 발성 속도, 발성기관의 형태와 기능, 피치 톤, 언어 습관 및 성별에 따라 다르게 나타난다. 목소리는 사람의 의사소통 핵심 요소이다. 제 4 차 산업 혁명의 시대에 목소리는 사람과 사람, 사람과 기계, 기계 와 기계 사이의 주요한 의사소통 수단이 된다. 그 이유 때문에 사람들은 자신의 의도를 다른 사람들에게 명확하게 전달하려고 노력한다. 그리고 이 과정에서 목소리는 언어 정보와 함께 다양한 추가 정보가 포함되게 된다. 예를 들어 감정 상태, 건강 상태, 신뢰도와 관련되거나, 거짓말의 여부, 음주로 인한 목소리의 변화 등 다양한 언어 및 비언어 정보를 포함하며, 다양한 분석 파라미터로 나타나게 된다. 이를 활용하면 개인의 신용도를 평가하는 척도로 사용할 수 있다. 특히 성대의 기본 주파수의 특성과 성도의 공진 주파수 특성의 관계를 분석함으로써 얻을 수 있다. 이전의 연구에서 다양한 신용 상태의 변화에 따른 목소리 분석 및 특성 변화를 연구 하였다. 본 연구에서는 음성을 통해 추출 된 매개 변수를 통해 기계 학습을 통한 개인 신용 판별 기를 제안한다.
The human voice is one of the easiest methods for the information transmission between human beings. The characteristics of voice can vary from person to person and include the speed of speech, the form and function of the vocal organ, the pitch tone, speech habits, and gender. The human voice is a ...
The human voice is one of the easiest methods for the information transmission between human beings. The characteristics of voice can vary from person to person and include the speed of speech, the form and function of the vocal organ, the pitch tone, speech habits, and gender. The human voice is a key element of human communication. In the days of the Fourth Industrial Revolution, voices are also a major means of communication between humans and humans, between humans and machines, machines and machines. And for that reason, people are trying to communicate their intentions to others clearly. And in the process, it contains various additional information along with the linguistic information. The Information such as emotional status, health status, part of trust, presence of a lie, change due to drinking, etc. These linguistic and non-linguistic information can be used as a device for evaluating the individual's credit worthiness by appearing in various parameters through voice analysis. Especially, it can be obtained by analyzing the relationship between the characteristics of the fundamental frequency(basic tonality) of the vocal cords, and the characteristics of the resonance frequency of the vocal track.In the previous research, the necessity of various methods of credit evaluation and the characteristic change of the voice according to the change of credit status were studied. In this study, we propose a personal credit discriminator by machine learning through parameters extracted through voice.
The human voice is one of the easiest methods for the information transmission between human beings. The characteristics of voice can vary from person to person and include the speed of speech, the form and function of the vocal organ, the pitch tone, speech habits, and gender. The human voice is a key element of human communication. In the days of the Fourth Industrial Revolution, voices are also a major means of communication between humans and humans, between humans and machines, machines and machines. And for that reason, people are trying to communicate their intentions to others clearly. And in the process, it contains various additional information along with the linguistic information. The Information such as emotional status, health status, part of trust, presence of a lie, change due to drinking, etc. These linguistic and non-linguistic information can be used as a device for evaluating the individual's credit worthiness by appearing in various parameters through voice analysis. Especially, it can be obtained by analyzing the relationship between the characteristics of the fundamental frequency(basic tonality) of the vocal cords, and the characteristics of the resonance frequency of the vocal track.In the previous research, the necessity of various methods of credit evaluation and the characteristic change of the voice according to the change of credit status were studied. In this study, we propose a personal credit discriminator by machine learning through parameters extracted through voice.
* AI 자동 식별 결과로 적합하지 않은 문장이 있을 수 있으니, 이용에 유의하시기 바랍니다.
가설 설정
In Fig. 3 (b), the change in pitch is relatively high and the variance is expected to be high.
제안 방법
At this time, the voices are created together with the linguistic communication, information, and the individual characteristics of the person who created it. In the process of making the voices, similar results are obtained by analyzing the characteristics of the vocal organs corresponding to the characteristics of the individual, the spatial, which is place who are live, characteristics of the language, the personality, the health, and the psychological state. Analyzing the voices, similar features remain in the form of sounds with meaning by language, and similar features are caused by different causes for individuals [1].
Voice analysis parameters can be extracted through cepstrum analysis and the parameters generated here can be used to improve machine performance through machine learning. In this study, we used actual voice data of people who had changed credit status, though few in number, and we performed a part to compare features. In the future, we will do more research to improve performance, make judgments, and save time using more data related to credit in the future.
The analysis is the result of the cepstrum analysis of the voice, the parameter for the resonance of the vocal track is confirmed in the low quefrency, and the information about the fundamental frequency of the vocal cord in the high quefrency.
In order to obtain parameter for this study, we were allowed to use the telephone voice record files for research purposes, which is a loan borrower and counselor from Korean savings banks(S-Capital and m-bank). Then, the recorded voices were analyzed through the parameterization, and the results from the analysis were compared with the actual personal credit score to find the similarities between these two. The 30 voices from the telephone conversation, comprising 18 males and 12 females, were collected.
대상 데이터
Then, the recorded voices were analyzed through the parameterization, and the results from the analysis were compared with the actual personal credit score to find the similarities between these two. The 30 voices from the telephone conversation, comprising 18 males and 12 females, were collected. The age group for the sample data is all in the 20s-40s.
이론/모형
However, machine learning method of deep learning requires lots of data, computing power to make a conclusion, and ultimately the resources to judge the result [1][3][4]. In this study, we use the support vector machine method which can produce effective results in small learning operations.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.