• The Journal of the Acoustical Society of Korea
• /
• v.18 no.5
• /
• pp.21-27
• /
• 1999

In this paper, we propose an improved MLP-based English pronunciation dictionary generator to apply to the variable vocabulary word recognizer. The variable vocabulary word recognizer can process any words specified in Korean word lexicon dynamically determined according to the current recognition task. To extend the ability of the system to task for English words, it is necessary to build a pronunciation dictionary generator to be able to process words not included in a predefined lexicon, such as proper nouns. In order to build the English pronunciation dictionary generator, we use context-dependent grapheme-to-phoneme multi-layer perceptron(MLP) architecture for each grapheme. To train each MLP, it is necessary to obtain grapheme-to-phoneme training data from general pronunciation dictionary. To automate the process, we use dynamic programming(DP) algorithm with some distance metrics. For training and testing the grapheme-to-phoneme MLPs, we use general English pronunciation dictionary with about 110 thousand words. With 26 MLPs each having 30 to 50 hidden nodes and the exception grapheme lexicon, we obtained the word accuracy of 72.8% for the 110 thousand words superior to rule-based method showing the word accuracy of 24.0%.

• Phonetics and Speech Sciences
• /
• v.12 no.4
• /
• pp.81-90
• /
• 2020

This paper describes a method of building Korean conversational speech data in the emergency medical domain and proposes an annotation method for the collected data in order to improve speech recognition performance. To suggest future research directions, baseline speech recognition experiments were conducted by using partial data that were collected and annotated. All voices were recorded at 16-bit resolution at 16 kHz sampling rate. A total of 166 conversations were collected, amounting to 8 hours and 35 minutes. Various information was manually transcribed such as orthography, pronunciation, dialect, noise, and medical information using Praat. Baseline speech recognition experiments were used to depict problems related to speech recognition in the emergency medical domain. The Korean conversational speech data presented in this paper are first-stage data in the emergency medical domain and are expected to be used as training data for developing conversational systems for emergency medical applications.

• Journal of Korea Multimedia Society
• /
• v.13 no.10
• /
• pp.1436-1443
• /
• 2010

In vocabulary recognition system has it's bad points of processing vocabulary unseen triphone and then no got distribution of confidence measure by cannot normalization. According to this problem to improve suggested VCOR(Version Control for Out-of Rejection) system by out-of vocabulary rejection algorithm use vocabulary management optimization and then phone data search support. In VCOR system to provide vocabulary information efficiently offering for user's vocabulary information using extend facet classification that improved for vocabulary measure management function offering accuracy of recognition for vocabulary. In this paper proposed system performance as a result of represent vocabulary dependence recognition rate of 97.56%, vocabulary independence recognition rate of 96.23%.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2015.07a
• /
• pp.155-158
• /
• 2015

단어의 유사도 측정 알고리즘은 DB 인덱싱, 필터링, 소스코드 분석 소프트웨어, 음성 인식 등 다양한 분야에서 활용되고 있다. 하지만 기존의 단어의 유사도만 비교하는 시스템에는 발음이 비슷한 유사단어나 오타가 있는 유사단어들은 측정을 못하는 단점이 있다. 언어의 유사도 측정에서는 알파벳만으로 볼게 아니라 언어 발음의 발화적 특성 또한 고려되어야 한다. 본 논문에서는 글로벌 시장에서의 다국적 기업들의 제품이나 문화 수출 등의 도움이 되는 각 나라의 금기어와의 발화적 특성까지 고려한 단어 유사도를 측정 할 수 있는 시스템을 제안한다. 11개국의 4개 언어 총 21487개의 금기어 단어를 금기어 데이터로 사용하였다. 제안하는 방법의 성능을 평가하기 위하여 타 알고리즘과의 성능비교와 여러 나라의 다양한 언어의 사용자들로부터 사용자 평가를 수행하였고 제안하는 방법이 발음 유사도를 측정하지 않는 알고리즘보다 우수한 성능을 보임을 확인하였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.05a
• /
• pp.447-450
• /
• 2019

본 논문은 한국어 문장의 감성 분류를 위해 문장의 형태소, 음절, 자소를 입력으로 하는 합성곱층과 DenseNet 을 적용한 Text Multi-channel DenseNet 모델을 제안한다. 맞춤법 오류, 음소나 음절의 축약과 탈락, 은어나 비속어의 남용, 의태어 사용 등 문법적 규칙에 어긋나는 다양한 표현으로 인해 단어 기반 CNN 으로 추출 할 수 없는 특징들을 음절이나 자소에서 추출 할 수 있다. 한국어 감성분석에 형태소 기반 CNN 이 많이 쓰이고 있으나, 본 논문에서 제안한 Text Multi-channel DenseNet 모델은 형태소, 음절, 자소를 동시에 고려하고, DenseNet 에 정보를 밀집 전달하여 문장의 감성 분류의 정확도를 개선하였다. 네이버 영화 리뷰 데이터를 대상으로 실험한 결과 제안 모델은 85.96%의 정확도를 보여 Multi-channel CNN 에 비해 1.45% 더 정확하게 문장의 감성을 분류하였다.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.10
• /
• pp.37-43
• /
• 2021

Using the acoustic features of speech, important social and linguistic information about the speaker can be obtained, and one of the key features is the dialect. A speaker's use of a dialect is a major barrier to interaction with a computer. Dialects can be distinguished at various levels such as phonemes, syllables, words, phrases, and sentences, but it is difficult to distinguish dialects by identifying them one by one. Therefore, in this paper, we propose a lightweight Korean dialect classification model using only MFCC among the features of speech data. We study the optimal method to utilize MFCC features through Korean conversational voice data, and compare the classification performance of five Korean dialects in Gyeonggi/Seoul, Gangwon, Chungcheong, Jeolla, and Gyeongsang in eight machine learning and deep learning classification models. The performance of most classification models was improved by normalizing the MFCC, and the accuracy was improved by 1.07% and F1-score by 2.04% compared to the best performance of the classification model before normalizing the MFCC.

• Phonetics and Speech Sciences
• /
• v.12 no.1
• /
• pp.85-95
• /
• 2020

Sentence repetition is a way of evaluating speech sound production to improve the limitation of word tests and spontaneous speech analysis. Speech sounds produced by children can be evaluated using several indicators. This study examined the progression of the percentage of correct consonants-revised (PCC-R) and phonological whole-word measure in different age and gender groups after setting consonants in various vowel contexts and implementing sentence repetition tasks that were designed to give all phonemes the chance to appear at least three times. For this study, 11 sentence repetition tasks were applied to 535 children aged 3 to 7 across the country, after which the resulting PCC-R and whole-word measure were analyzed. The study results showed that all the indicators improved in older age groups and there were significant differences depending on age, however, no significant differences dependent on gender were found. The sentence repetition conditions data used in this study were collected from across the country, and the age difference between each age group was six months. This study is noteworthy because it collected a sufficient amount of data from each group, highlighted the limitation of the word naming and the spontaneous speech analysis, and suggests new criteria of evaluation through the analysis of each whole-word measure in sentence repetition, which was not applied in previous studies.

• The Journal of the Acoustical Society of Korea
• /
• v.40 no.5
• /
• pp.530-536
• /
• 2021

Recently, due to the development of deep learning, end-to-end speech recognition, which directly maps graphemes to speech signals, shows good performance. Especially, among the end-to-end models, conformer shows the best performance. However end-to-end models only focuses on the probability of which grapheme will appear at the time. The decoding process uses a greedy search or beam search. This decoding method is easily affected by the final probability output by the model. In addition, the end-to-end models cannot use external pronunciation and language information due to structual problem. Therefore, in this paper conformer with lexicon transducer is proposed. We compare phoneme-based model with lexicon transducer and grapheme-based model with beam search. Test set is consist of words that do not appear in training data. The grapheme-based conformer with beam search shows 3.8 % of CER. The phoneme-based conformer with lexicon transducer shows 3.4 % of CER.

• Phonetics and Speech Sciences
• /
• v.11 no.1
• /
• pp.41-49
• /
• 2019

An end-to-end speech recognition model consisting of a single integrated neural network model was recently proposed. The end-to-end model does not need several training steps, and its structure is easy to understand. However, it is difficult to understand how the model recognizes speech internally. In this paper, we visualized and analyzed the attention-based end-to-end model to elucidate its internal mechanisms. We compared the acoustic model of the BLSTM-HMM hybrid model with the encoder of the end-to-end model, and visualized them using t-SNE to examine the difference between neural network layers. As a result, we were able to delineate the difference between the acoustic model and the end-to-end model encoder. Additionally, we analyzed the decoder of the end-to-end model from a language model perspective. Finally, we found that improving end-to-end model decoder is necessary to yield higher performance.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.3
• /
• pp.83-90
• /
• 2024

In this paper, we propose a solution to apply a proximity voice dialog system to voice dialog technology, one of the interaction systems in multimedia environments. A voice dialog between multiple users in a multimedia space is designed by adjusting the volume of the voice according to the distance between the user avatars and muting the user who is beyond the audible distance. The main feature of this research is a reliable UDP-based active server system that delivers low-quality voice data to users who are far away based on distance and does not transmit voice data to users who enter the inaudible area for economic development. The performance of the proposed system was measured in a previously completed project based on the Unity game engine, and it is expected that the system proposed in this research will be actively used in environments that provide interaction between multiple users such as met averse content and real-time battle action games.