• Proceedings of the KSPS conference
• /
• 2003.10a
• /
• pp.69-72
• /
• 2003

In this paper, we investigate various factors that are relevant to design and implementation of an integrated management system for various speech corpora. The purpose of this paper is to manage an integrated management system for various kinds of speech corpora necessary for speech research and speech corpora consrtructed in different data formats. In addition, ways are considered to allow users to search with effect for speech corpora that meet various conditions which they want, and to allow them to add with ease corpora that are constructed newly. In order to achieve this goal, we design a global schema for an integrated management of new additional information without changing old speech corpora, and construct a web-based integrated management system based on the scheme that can be accessed without any temporal and spatial restrictions. And we show the steps by which these can be implemented, and describe related future study topics, examining the system.

• Journal of Multimedia Information System
• /
• v.7 no.1
• /
• pp.33-40
• /
• 2020

Machine learning consists of supervised and unsupervised learning among which supervised learning is used for the speech recognition objectives. Supervised learning is the Data mining task of inferring a function from labeled training data. Speech recognition is the current trend that has gained focus over the decades. Most automation technologies use speech and speech recognition for various perspectives. This paper demonstrates an overview of major technological standpoint and gratitude of the elementary development of speech recognition and provides impression method has been developed in every stage of speech recognition using supervised learning. The project will use DNN to recognize speeches using magnitudes with large datasets.

• Speech Sciences
• /
• v.11 no.2
• /
• pp.247-257
• /
• 2004

There has been many research efforts to overcome the problems of speech recognition in the noisy conditions. Among them, the model-based compensation methods such as the parallel model combination (PMC) and vector Taylor series (VTS) have been found to perform efficiently compared with the previous speech enhancement methods or the feature-based approaches. In this paper, a data-driven model compensation approach that adapts the HMM(hidden Markv model) parameters for the noisy speech recognition is proposed. Instead of assuming some statistical approximations as in the conventional model-based methods such as the PMC, the statistics necessary for the HMM parameter adaptation is directly estimated by using the Baum-Welch algorithm. The proposed method has shown improved results compared with the PMC for the noisy speech recognition.

• Speech Sciences
• /
• v.13 no.1
• /
• pp.129-139
• /
• 2006

In this paper, we propose two effective energy feature normalization methods for robust speech recognition in noisy environments. In the first method, we estimate the noise energy and remove it from the noisy speech energy. In the second method, we propose a modified algorithm for the Log-energy Dynamic Range Normalization (ERN) method. In the ERN method, the log energy of the training data in a clean environment is transformed into the log energy in noisy environments. If the minimum log energy of the test data is outside of a pre-defined range, the log energy of the test data is also transformed. Since the ERN method has several weaknesses, we propose a modified transform scheme designed to reduce the residual mismatch that it produces. In the evaluation conducted on the Aurora2.0 database, we obtained a significant performance improvement.

• Proceedings of the Korean Information Science Society Conference
• /
• 2006.10d
• /
• pp.302-305
• /
• 2006

In this paper, we have proposed a novel method using quantization of wavelet coefficients for secret data communication. First, speech signal is partitioned into small time frames and the frames are transformed into frequency domain using a WT(Wavelet Transform). We quantize the wavelet coefficients and embedded secret data into the quantized wavelet coefficients. The destination regard quantization errors of received speech as seceret dat. As most speech watermark techniques have a trade off between noise robustness and speech quality, our method also have. However we solve the problem with a partial quantization and a noise level dependent threshold. In additional, we improve the speech quality with de-noising method using wavelet transform. Since the signal is processed in the wavelet domain, we can easily adapt the de-noising method based on wavelet transform. Simulation results in the various noisy environments show that the proposed method is reliable for secret communication.

• The Journal of the Acoustical Society of Korea
• /
• v.20 no.3
• /
• pp.56-62
• /
• 2001

In this paper, we adopt a missing data theory to speech recognition. It can be used in order to maintain high performance of speech recognizer when the missing data occurs. In general, hidden Markov model (HMM) is used as a stochastic classifier for speech recognition task. Acoustic events are represented by continuous probability density function in continuous density HMM(CDHMM). The missing data theory has an advantage that can be easily applicable to this CDHMM. A marginalization method is used for processing missing data because it has small complexity and is easy to apply to automatic speech recognition (ASR). Also, a spectral subtraction is used for detecting missing data. If the difference between the energy of speech and that of background noise is below given threshold value, we determine that missing has occurred. We propose a new method that examines the reliability of detected missing data using voicing probability. The voicing probability is used to find voiced frames. It is used to process the missing data in voiced region that has more redundant information than consonants. The experimental results showed that our method improves performance than baseline system that uses spectral subtraction method only. In 452 words isolated word recognition experiment, the proposed method using the voicing probability reduced the average word error rate by 12％ in a typical noise situation.

• 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.

• Phonetics and Speech Sciences
• /
• v.4 no.4
• /
• pp.19-27
• /
• 2012

When two speakers read a text together, the produced speech has been shown to reduce a high degree of variability (e.g., pause duration and placement, and speech rate). This paper provides a quantitative analysis of speech rate variation exhibited in synchronous speech by examining the global and local patterns in two dialects of Mandarin Chinese (Taiwan and Shanghai). We analyzed the speech data in terms of mean speech rate and the reference of "Just Noticeable difference (JND)" within a subject and across subjects. Our findings show that speakers show lower and less variable speech rates when they read a text synchronously than when they read alone. This global pattern is observed consistently across speakers and dialects maintaining the unique local variation patterns of speech rate for each dialect. We conclude that paired speakers lower their speech rates and decrease the variability in order to ensure the synchrony of their speech.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.3
• /
• pp.69-74
• /
• 2022

SNN (Spiking Neural Network) operating in neuromorphic architecture was created by mimicking human neural networks. Neuromorphic computing based on neuromorphic architecture requires relatively lower power than typical deep learning techniques based on GPUs. For this reason, research to support various artificial intelligence models using neuromorphic architecture is actively taking place. This paper conducted a performance analysis of the speech recognition model based on neuromorphic architecture according to the speech data preprocessing technique. As a result of the experiment, it showed up to 84% of speech recognition accuracy performance when preprocessing speech data using the Fourier transform. Therefore, it was confirmed that the speech recognition service based on the neuromorphic architecture can be effectively utilized.

• Phonetics and Speech Sciences
• /
• v.12 no.2
• /
• pp.29-37
• /
• 2020

Recently, the neural network-based deep learning algorithm has dramatically improved performance compared to the classical Gaussian mixture model based hidden Markov model (GMM-HMM) automatic speech recognition (ASR) system. In addition, researches on end-to-end (E2E) speech recognition systems integrating language modeling and decoding processes have been actively conducted to better utilize the advantages of deep learning techniques. In general, E2E ASR systems consist of multiple layers of encoder-decoder structure with attention. Therefore, E2E ASR systems require data with a large amount of speech-text paired data in order to achieve good performance. Obtaining speech-text paired data requires a lot of human labor and time, and is a high barrier to building E2E ASR system. Therefore, there are previous studies that improve the performance of E2E ASR system using relatively small amount of speech-text paired data, but most studies have been conducted by using only speech-only data or text-only data. In this study, we proposed a semi-supervised training method that enables E2E ASR system to perform well in corpus in different domains by using both speech or text only data. The proposed method works effectively by adapting to different domains, showing good performance in the target domain and not degrading much in the source domain.