• Journal of Information Processing Systems
• /
• v.11 no.1
• /
• pp.22-38
• /
• 2015

While Modern Standard Arabic is the formal spoken and written language of the Arab world; dialects are the major communication mode for everyday life. Therefore, identifying a speaker's dialect is critical in the Arabic-speaking world for speech processing tasks, such as automatic speech recognition or identification. In this paper, we examine two approaches that reduce the Universal Background Model (UBM) in the automatic dialect identification system across the five following Arabic Maghreb dialects: Moroccan, Tunisian, and 3 dialects of the western (Oranian), central (Algiersian), and eastern (Constantinian) regions of Algeria. We applied our approaches to the Maghreb dialect detection domain that contains a collection of 10-second utterances and we compared the performance precision gained against the dialect samples from a baseline GMM-UBM system and the ones from our own improved GMM-UBM system that uses a Reduced UBM algorithm. Our experiments show that our approaches significantly improve identification performance over purely acoustic features with an identification rate of 80.49%.

• Journal of Intelligence and Information Systems
• /
• v.27 no.2
• /
• pp.17-32
• /
• 2021

Speaker recognition is generally divided into speaker identification and speaker verification. Speaker recognition plays an important function in the automatic voice system, and the importance of speaker recognition technology is becoming more prominent as the recent development of portable devices, voice technology, and audio content fields continue to expand. Previous speaker recognition studies have been conducted with the goal of automatically determining who the speaker is based on voice files and improving accuracy. Speech is an important sociolinguistic subject, and it contains very useful information that reveals the speaker's attitude, conversation intention, and personality, and this can be an important clue to speaker recognition. The final ending used in the speaker's speech determines the type of sentence or has functions and information such as the speaker's intention, psychological attitude, or relationship to the listener. The use of the terminating ending has various probabilities depending on the characteristics of the speaker, so the type and distribution of the terminating ending of a specific unidentified speaker will be helpful in recognizing the speaker. However, there have been few studies that considered speech in the existing text-based speaker recognition, and if speech information is added to the speech signal-based speaker recognition technique, the accuracy of speaker recognition can be further improved. Hence, the purpose of this paper is to propose a novel method using speech style expressed as a sentence-final ending to improve the accuracy of Korean speaker recognition. To this end, a method called sentence sequencing that generates vector values by using the type and frequency of the sentence-final ending appearing in the utterance of a specific person is proposed. To evaluate the performance of the proposed method, learning and performance evaluation were conducted with a actual drama script. The method proposed in this study can be used as a means to improve the performance of Korean speech recognition service.

• 제어로봇시스템학회:학술대회논문집
• /
• 2002.10a
• /
• pp.75.5-75
• /
• 2002

$\textbullet$ FFT $\textbullet$ Autocorrelation $\textbullet$ Levinson_Durbin resolution $\textbullet$ LP coefficients $\textbullet$ LP cepstral Coefficients $\textbullet$ Incremental Learning

• Journal of Internet Computing and Services
• /
• v.22 no.6
• /
• pp.33-40
• /
• 2021

Automatic Speech Recognition(ASR) is a technology that analyzes human speech sound into speech signals and then automatically converts them into character strings that can be understandable by human. Speech recognition technology has evolved from the basic level of recognizing a single word to the advanced level of recognizing sentences consisting of multiple words. In real-time voice conversation, the high recognition rate improves the convenience of natural information delivery and expands the scope of voice-based applications. On the other hand, with the active application of speech recognition technology, concerns about related cyber attacks and threats are also increasing. According to the existing studies, researches on the technology development itself, such as the design of the Automatic Speaker Verification(ASV) technique and improvement of accuracy, are being actively conducted. However, there are not many analysis studies of attacks and threats in depth and variety. In this study, we propose a cyber attack model that bypasses voice authentication by simply manipulating voice frequency and voice speed for AI voice recognition service equipped with automated identification technology and analyze cyber threats by conducting extensive experiments on the automated identification system of commercial smartphones. Through this, we intend to inform the seriousness of the related cyber threats and raise interests in research on effective countermeasures.

• Phonetics and Speech Sciences
• /
• v.9 no.2
• /
• pp.9-22
• /
• 2017

The purpose of this paper is to report the whole process of developing Korean Standard Speech Database (KSS DB). This project is supported by SPO (Supreme Prosecutors' Office) research grant for three years from 2014 to 2016. KSS DB is designed to provide speech data for acoustic-phonetic and phonological studies and speaker recognition system. For the samples to represent the spoken Korean, sociolinguistic factors, such as region (9 regional dialects), age (5 age groups over 20) and gender (male and female) were considered. The goal of the project is to collect over 3,000 male and female speakers of nine regional dialects and five age groups employing direct and indirect methods. Speech samples of 3,191 speakers (2,829 speakers and 362 speakers using direct and indirect methods, respectively) are collected and databased. KSS DB designs to collect read and spontaneous speech samples from each speaker carrying out 5 speech tasks: three (pseudo-)spontaneous speech tasks (producing prolonged simple vowels, 28 blanked sentences and spontaneous talk) and two read speech tasks (reading 55 phonetically and phonologically rich sentences and reading three short passages). KSS DB includes a 16-bit, 44.1kHz speech waveform file and a orthographic file for each speech task.

• MALSORI
• /
• no.41
• /
• pp.19-30
• /
• 2001

As a pan of voice identification, this study is on the voices from four different groups of sisters. The purpose of this study is to observe and analyze if sister's voices are actually so similar, as many people believe, that it is more difficult to distinguish one from the other than those of unrelated speakers. And it is also aimed to find out what the clues are to decide whether or not the voices are those of the same speaker or of different ones. The result shows that the mean frequencies of $F_3$ and $F_4$ of examined vowels and F0, in some group of sisters, are analyzed similar, In aural test, many listeners took their voices for those of the same speaker.

• Proceedings of the Korean Society for Language and Information Conference
• /
• 2007.11a
• /
• pp.311-320
• /
• 2007

Interrogative sentences are generally used to perform speech acts of directly asking a question or making a request, but they are also used to convey such speech acts indirectly. In the utterances, such indirect uses of interrogative sentences usually carry speaker's emotion with a negative attitude, which is close to an expression of anger. The identification of such negative emotion is known as a difficult problem that requires relevant information in syntax, semantics, discourse, pragmatics, and speech signals. In this paper, we argue that the interrogatives used for indirect speech acts could serve as a dominant marker for identifying the emotional attitudes, such as anger, as compared to other emotion-related markers, such as discourse markers, adverbial words, and syntactic markers. To support such an argument, we analyze the dialogues collected from the Korean soap operas, and examine individual or cooperative influences of the emotion-related markers on emotional realization. The user study shows that the interrogatives could be utilized as a promising device for emotion identification.

• The Journal of the Acoustical Society of Korea
• /
• v.38 no.5
• /
• pp.593-600
• /
• 2019

The similarity in tones between speakers can lower the performance of speaker verification. To improve the performance of speaker verification systems, we propose a multi-task learning technique using deep neural network to learn speaker information and age information. Multi-task learning can improve generalization performances, because it helps deep neural networks to prevent hidden layers from overfitting into one task. However, we found in experiments that learning of age information does not work well in the process of learning the deep neural network. In order to improve the learning, we propose a method to dynamically change the objective function weights of speaker identification and age estimation in the learning process. Results show the equal error rate based on RSR2015 evaluation data set, 6.91 % for the speaker verification system without using age information, 6.77 % using age information only, and 4.73 % using age information when weight change technique was applied.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1993.06a
• /
• pp.121-124
• /
• 1993

This paper is a study on the word recognition using neural network. A limited vocabulary, speaker independent, isolated word recognition system has been built. This system recognizes isolated word without performing segmentation, phoneme identification, or dynamic time wrapping. It needs a static pattern approach to recognize a spatio-temporal pattern. The preprocessing only includes preceding and tailing silence removal, and word length determination. A LPC analysis is performed on each of 24 equally spaced frames. The PARCOR coefficients plus 3 other features from each frame is extracted. In order to simplify a structure of neural network, we composed binary code form to decrease output nodes.

• Proceedings of the IEEK Conference
• /
• 2001.09a
• /
• pp.793-796
• /
• 2001

음성인식 시스템은 인간의 의사소통 수단인 음성을 기계가 인지할 수 있게 하는 것이다. 이러한 음성 인식 알고리즘 개발은 현재 활발히 진행되고 있다. 올바른 음성인식 시스템의 구현을 위해서는 높은 인식률 구현과 적은 처리시간이 요구된다. 또한 인식률 향상을 위해서는 그 구현 알고리즘이 복잡해지고 이에 따라 많은 처리 시간이 요구된다. 본 논문에서는 성문 특성에 따른 Glottal Spectrum에 적응적인 필터계수를 적용하여 인식률 향상을 도모하였다. 제안한 알고리즘을 모의 실험한 결과 전체 인식률이 2% 향상되었다.