• 말소리와 음성과학
• /
• 제15권4호
• /
• pp.1-10
• /
• 2023

본 연구에서는 AI Hub에 구축된 '교육용 한국인의 영어 음성 데이터'에 있는 발음 평가 데이터를 활용하여 철자 기반 발화 속도 및 조음 속도와 음절 기반 발화 속도 및 조음 속도 중 발음 정확성 및 운율 유창성, 합산 점수를 예측하는 모델에 어떤 요소가 더 유의미한 영향을 미치는지 분석하였다. 이를 위해 13세, 19세, 26세 연령별, 성별, 수준별로 이 코퍼스의 훈련 데이터에서 총 900개 발화를 추출하여 데이터에 포함된 다양한 요소를 활용해 평가 점수를 예측하는 선형효과분석을 실행하였다. 선형효과분석에서 최적의 세 개 모델을 통해 예측된 평가 점수를 검증 데이터에서 추출한 총 180개 발화의 평가 점수와 얼마나 상관관계가 있는지도 분석하였다. 분석 결과 발음의 정확성과 운율의 유창성, 합산 점수 예측 모델 모두 철자 기반 발화 속도와 조음 속도보다 음절 기반 발화 속도와 조음 속도가 평가 점수를 예측하는데 더 큰 영향을 주는 것으로 밝혀졌다. 모델에서 예측한 점수와 검정 데이터의 실제 점수와의 상관계수는 .65에서 .68 사이로 각 모델의 평가 점수 예측력이 나쁘지 않았다. 발화 속도와 조음 속도 간에 어떤 요소가 더 큰 영향을 미치는지는 본 연구를 통해 밝혀내지 못하였다.

• 전자공학회논문지B
• /
• 제33B권4호
• /
• pp.131-140
• /
• 1996

In this paper, we propose the method that extracts the speech feature using the hearing model through signal precessing techniques. The proposed method includes following procedure ; normalization of the short-time speech block by its maximum value, multi-resolution analysis using the discrete wavelet transformation and re-synthesize using thediscrete inverse wavelet transformation, differentiation after analysis and synthesis, full wave rectification and integration. In order to verify the performance of the proposed speech feature in the speech recognition task, korean digita recognition experiments were carried out using both the dTW and the VQ-HMM. The results showed that, in case of using dTW, the recognition rates were 99.79% and 90.33% for speaker-dependent and speaker-independent task respectively and, in case of using VQ-HMM, the rate were 96.5% and 81.5% respectively. And it indicates that the proposed speech feature has the potentials to use as a simple and efficient feature for recognition task.

• 음성과학
• /
• 제14권4호
• /
• pp.131-144
• /
• 2007

In an effort to improve the performance of mel frequency cepstral coefficients (MFCC), we investigate the effects of varying the parameters for the filter banks and their associated windows on speech recognition rates. Specifically, the mel and bark scales are combined with various types of filter bank windows. Comparison and evaluation of the suggested methods are performed by two independent ways of speech recognition and the Fisher discriminant objective function. It is shown that the Hanning window based on the bark scale yields 28.1% relative performance improvements over the triangular window with the mel scale in speech recognition error rate. Further work on incorporating PCA and/or LDA would be desirable as a postprocessor to MFCC extraction.

• 말소리와 음성과학
• /
• 제4권1호
• /
• pp.29-47
• /
• 2012

The purpose of this paper is to show how to convert the label files of the Buckeye Corpus of Spontaneous Speech [1] into Praat format and to introduce some of the Praat scripts that will enable linguists to study various aspects of spoken American English present in the corpus. During the conversion process, several types of errors were identified and corrected either manually or automatically by the use of scripts. The Praat script tools that have been developed can help extract from the corpus massive amounts of phonetic measures such as the VOT of plosives, the formants of vowels, word frequency information and speech rates that span several consecutive words. The script tools can extract additional information concerning the phonetic environment of the target words or allophones.

• ETRI Journal
• /
• 제46권1호
• /
• pp.71-81
• /
• 2024

Recent advances in deep learning for speech and visual recognition have accelerated the development of multimodal speech recognition, yielding many innovative results. We introduce a Korean audiovisual speech recognition corpus. This dataset comprises approximately 150 h of manually transcribed and annotated audiovisual data supplemented with additional 2000 h of untranscribed videos collected from YouTube under the Creative Commons License. The dataset is intended to be freely accessible for unrestricted research purposes. Along with the corpus, we propose an open-source framework for automatic speech recognition (ASR) and audiovisual speech recognition (AVSR). We validate the effectiveness of the corpus with evaluations using state-of-the-art ASR and AVSR techniques, capitalizing on both pretrained models and fine-tuning processes. After fine-tuning, ASR and AVSR achieve character error rates of 11.1% and 18.9%, respectively. This error difference highlights the need for improvement in AVSR techniques. We expect that our corpus will be an instrumental resource to support improvements in AVSR.

• 한국음향학회:학술대회논문집
• /
• 한국음향학회 1994년도 제11회 음성통신 및 신호처리 워크샵 논문집 (SCAS 11권 1호)
• /
• pp.199-204
• /
• 1994

In this paper, we researched abut CELP speech coding algorithm using efficlent pseudo-stochastic block codes, adaptive-codebook and improved fixed-gain codebook. The pseudo-stochastic block codes refer to stochastically populated block codes in which the adjacent codewords in an innovation codebook are non-independent. The adaptive-codebook was made with previous prediction speech data by storage-shift register. This CELP coding algorithm enables the coding of toll quality speech at bit rates from 4.8kbits/s to 9.6 kbits/s. This algorithm was realized TMS320C30 microprocessor in realtime.

• 대한전기학회논문지
• /
• 제39권6호
• /
• pp.607-615
• /
• 1990

In this paper, I propose the experimental models to classify the consonants in the Vowel-Consonant-Vowel (VCV) speech chain into four phonemic groups such as nasals, liquids, plosives and the others. To classify the fuzzy patterns like speech, it is necessary to analyze the distribution of acoustic feature of many training data. The classification rules are maximum 4 th order polynomial functions obtained by regression analysis, contributing collectively the result. The final result shows about 87% success rates with the data spoken by one man.

• 한국정보통신학회논문지
• /
• 제5권3호
• /
• pp.451-458
• /
• 2001

본 연구는 자동차내의 각종 제어장치들을 음성으로 실시간 제어하기 위한 음성제어 시스템을 제안하고 실험적으로 검증하였다. 실시간 제어음성 인식시스템은 8bit-l0MHz로 A/D변환된 음성 데이터를 실시간으로 시작점과 끝점을 검출한 후, One Pass DP법으로 인식하였으며 그 결과를 모니터에 문장으로 출력하며 제어용 인터페이스에 제어데이터를 보내도록 구성하였다. HMM모델은 자동차내의 장치들을 제어하기 위한 제어음성 및 숫자음들로 구성되는 연속음성을 학습 및 모델링 하였다. 단어.제어문들의 인식률은 평균 97.3%, 숫자음의 경우는 평균 96.3% 정도의 인식률을 얻을 수 있었다.

• 음성과학
• /
• 제13권1호
• /
• pp.69-83
• /
• 2006

Speech characteristics of Wernicke's aphasia are characterized by such errors as empty speech, jargon, paraphasia, filler and others. However, not all the errors can be observed in each patient presumably due to diverse auditory comprehension (AC) abilities and/or lesion loci. The purpose of this study was, thus, to clarify the speech characteristics of Wernicke's aphasics according to the AC levels (i.e., better vs. worse) and lesion loci (i.e., Wernicke's area, WA vs. non-Wernicke's area, NWA). The authors divided 21 Wernicke's aphasic patients into four patient groups based on their AC levels and the lesion loci. The results showed that the four groups differed only in CIU (Correct Information Unit) rate. The patient groups with a better AC ability had higher CIU rates than the groups with a worse AC regardless of the lesion loci (e.g., WA or NWA). Therefore, it was concluded that CIU rate, the differentiating speech variable was most likely related to the AC levels, but not to lesion loci.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2000년도 하계학술대회 논문집 D
• /
• pp.3166-3168
• /
• 2000

This paper proposed variable threshold dual rate ADPCM coding method which is modified from the standard ADPCM of ITU G.726 for speech quality improvement. The speech quality of variable threshold dual rate ADPCM is better than single rate ADPCM at noisy environment without increasing the complexity by using ZCR(Zero Crossing Rate). In this case, ZCR is used to divide input signal samples into two categories(noisy & speech). The samples with higher ZCR is categorized as the noisy region and the samples with lower ZCR is categorized as the speech region. Noisy region uses higher threshold value to be compressed by 16Kbps for reduced bit rates and the speech region uses lower threshold value to be compressed by 40Kbps for improved speech quality. Comparing with the conventional ADPCM, which adapts the fixed coding rate. the proposed variable threshold dual rate ADPCM coding method improves noise character without increasing the bit rate. For real time applications, ZCR calculation was considered as a simple method to obtain the background noise information for preprocess of speech analysis such as FFT and the experiment showed that the simple calculation of ZCR can be used without complexity increase. Dual rate ADPCM can decrease the amount of transferred data efficiently without increasing complexity nor reducing speech quality. Therefore result of this paper can be applied for real-time speech application such as the internet phone or VoIP.