음향 특성에 따른 한국어 방언 분류 모델의 성능 비교
Performance Comparison of Korean Dialect Classification Models Based on Acoustic Features
2
조회
0
다운로드
1
인용
0
참고문헌
키워드
초록
말소리의 음향 특징을 이용하여 화자에 대한 중요한 사회, 언어학적 정보를 얻을 수 있는데 그중 한 가지 핵심 특징은 방언이다. 화자의 방언 사용은 컴퓨터와의 상호작용을 방해하는 주요 요소이다. 방언은 발화의 음소, 음절, 단어, 문장 및 구와 같이 다양한 수준에서 구분할 수 있지만이를 하나하나 식별하여 방언을 구분하기는 어렵다. 이에 본 논문에서는 음성 데이터의 특성 중MFCC만 사용하는 경량화된 한국어 방언 분류 모델을 제안한다. 한국인 대화 음성 데이터를 통해MFCC 특징을 활용하는 최적의 방법을 연구하고, 8가지 머신 러닝 및 딥러닝 분류 모델에서 경기/서울, 강원, 충청, 전라, 경상 5개의 한국어 방언 분류 성능을 비교한다. MFCC를 정규화하는 방법으로 대부분의 분류 모델에서 성능을 향상시켰으며, MFCC를 정규화하기 전 분류 모델의 최고성능과 비교하여 정확도는 1.07%, F1-score는 2.04% 향상된 성능을 기록하였다.
Abstract
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.
참고문헌 (22)
- "Deep Learning-based Speech Recognition Technology"[URL]
- S. S. Jo, "AI(Artificial Intelligence)Voice Assistant Evolving to Platform", IITP, pp.1~25 (2017)
- Rongqing Huang, "Dialect", accent classification via boosted word modeling, 2005, pp.1, I~585[DOI]
- Thomas Purnell, "Perceptual and phonetic experiments on american english dialect identification", Journal of language and social psychology, 18(1), pp.10~30 (1999)
- Najim Dehak, "Front-End Factor Analysis for Speaker Verification", IEEE Transactions on Audio, Speech, and Language Processing, 19(4), pp.788~798 (2011), Institute of Electrical and Electronics Engineers (IEEE)[DOI]
- Dehak, N., "Language recognition via ivectors and dimensionality reduction", Proceedings of the Annual Conference of the International Speech Communication Association, pp.857~860 (2011)
- Yan Song, "i‐vector representation based on bottleneck features for language identification", Electronics Letters, 49(24), pp.1569~1570 (2013), Institution of Engineering and Technology (IET)[DOI]
- Snyder, D., "Spoken language recognition using x-vectors", Proceedings of Odyssey 2018 The Speaker and Language Recognition Workshop, pp.105~111 (2018)
- Charalambos Themistocleous, "Dialect Classification From a Single Sonorant Sound Using Deep Neural Networks", Frontiers in Communication, 4 (2019), Frontiers Media SA[DOI]
- Nagaratna B. Chittaragi, "Acoustic features based word level dialect classification using SVM and ensemble methods", IEEE Trans, In 2017 Tenth International Conference on Contemporary Computing (IC3), pp.1~6 (2017)
- Li, Ming, "Combining five acoustic level modeling methods for automatic speaker age and gender recognition", Eleventh Annual Conference of the International Speech Communication Association, pp.2526~2829 (2010)
- Reynolds, Douglas A., "Speaker verification using adapted Gaussian mixture models", Digital signal processing, 10, pp.19~41 (2000)
- Li, Ming, "Spoken language identification using score vector modeling and support vector machine", Eighth Annual Conference of the International Speech Communication Association, pp.350~353 (2007)
- Stolcke, Andreas, "Speaker recognition with session variability normalization based on MLLR adaptation transforms", IEEE Transactions on Audio, Speech, and Language Processing, 15(7), pp.1987~1998 (2007)
- Qawaqneh, Zakariya, "Deep neural network framework and transformed MFCCs for speaker's age and gender classification", Knowledge-Based Systems, 115, pp.5~14 (2017)
- Mallouh, Arafat Abu, "New transformed features generated by deep bottleneck extractor and a GMM–UBM classifier for speaker age and gender classification", Neural Computing and Applications, 30(8), pp.2581~2593 (2018)
- Ghahremani, P., "End-to-end Deep Neural Network Age Estimation", INTERSPEECH, pp.277~281 (2018)
- Snyder, D., "X-vectors: Robust dnn embeddings for speaker recognition", 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp.5329~5333 (2018)
- Hourri, Soufiane, "A deep learning approach for speaker recognition", International Journal of Speech Technology, 23(1), pp.123~131 (2020)
- S. Gopal Krishna Patro, "Normalization:A Preprocessing Stage", arXiv preprint arXiv:1503.06462, ~ (2015)
- Goutte, Cyril, "A probabilistic interpretation of precision, recall and F-score, with implication for evaluation", European conference on information retrieval, pp.345~359 (2005), Springer
- "Korean Conversation Voice"[URL]