• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.38 no.5
• /
• pp.567-575
• /
• 2001

This paper suggests new method to get fixed size parameter from different length of voice signals. The efficiency of speech recognizer is determined by how to compare the similarity(distance of each pattern) of the parameter from voice signal. But the variation of voice signal and the difference of speech speed make it difficult to extract the fixed size parameter from the voice signal. The method suggested in this paper is to normalize the parameter at fixed size by using the 2 dimension DCT(Discrete Cosine Transform) after representing the parameter by spectrogram. To prove validity of the suggested method, parameter extracted from 32 auditory filter-bank(it estimates auditory nerve firing probabilities) is used for the input of neural network after being processed by 2 dimension DCT. And to compare with conventional methods, we used one of conventional methods which solve time alignment problem. The result shows more efficient performance and faster recognition speed in the speaker dependent and independent isolated word recognition than conventional method.

• The Journal of the Acoustical Society of Korea
• /
• v.18 no.2
• /
• pp.3-10
• /
• 1999

This paper describes the development and evaluation of a Korean address input system employing automatic speech recognition technique as user interface for input Korean address. Address consists of cities, provinces and counties. The system works on a window 95 environment of personal computer with built-in soundcard. In the speech recognition part, the Continuous density Hidden Markov Model(CHMM) for making phoneme like units(PLUs) and One Pass Dynamic Programming(OPDP) algorithm is used for recognition. For address recognition, Finite State Automata(FSA) suitable for Korean address structure is constructed. To achieve an acceptable performance against the variation of speakers, microphones, and environmental noises, Maximum a posteriori(MAP) estimation is implemented in adaptation. And to improve the recognition speed, fast search method using variable pruning threshold is newly proposed. In the evaluation tests conducted for the 100 connected words uttered by 3 males the system showed above average 96.0% of recognition accuracy for connected words after adaption and recognition speed within 2 seconds, showing the effectiveness of the system.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.39 no.2
• /
• pp.116-127
• /
• 2002

A new approach to improve the speaker adaptation algorithm by means of the variable number of observation density functions for CDHMM speech recognizer has been proposed. The proposed method uses the observation density function with more than one mixture in each state to represent speech characteristics in detail. The number of mixtures in each state is determined by the number of frames and the determinant of the variance, respectively. The each MAP Parameter is extracted in every mixture determined by these two methods. In addition, the state segmentation method requiring speaker adaptation can segment the adapting speech more Precisely by using speaker-independent model trained from sufficient database as a priori knowledge. And the state duration distribution is used lot adapting the speech duration information owing to speaker's utterance habit and speed. The recognition rate of the proposed methods are significantly higher than that of the conventional method using one mixture in each state.

• MALSORI
• /
• no.50
• /
• pp.51-69
• /
• 2004

In this paper, we present a one-pass semi-dynamic network decoding framework that inherits both advantages of fast decoding speed from static network decoders and memory efficiency from dynamic network decoders. Our method is based on the novel language model network representation that is essentially of finite state machine (FSM). The static network derived from the language model network [1][2] is partitioned into smaller subnetworks which are static by nature or self-structured. The whole network is dynamically managed so that those subnetworks required for decoding are cached in memory. The network is near-minimized by applying the tail-sharing algorithm. Our decoder is evaluated on the 25k-word Korean broadcast news transcription task. In case of the search network itself, the network is reduced by 73.4% from the tail-sharing algorithm. Compared with the equivalent static network decoder, the semi-dynamic network decoder has increased at most 6% in decoding time while it can be flexibly adapted to the various memory configurations, giving the minimal usage of 37.6% of the complete network size.

• The Journal of the Acoustical Society of Korea
• /
• v.18 no.7
• /
• pp.112-117
• /
• 1999

In case of spectral subtraction for noise robust speech recognition system, this method often makes loss of speech signal. In this study, we propose a method that variation and determination of Gaussian function at semi-continuous HMM(Hidden Markov Model) is made on the basis of SNR criteria function, in which SNR means signal to noise ratio between estimation noise and subtracted signal per frame. For proving effectiveness of this method, we show the estimation error to be related with the magnitude of estimated noise through signal waveform. For this reason, Gaussian function is varied and determined by SNR. When we test recognition rate by computer simulation under the noise environment of driving car over the speed of 80㎞/h, the proposed Gaussian decision method by SNR turns out to get more improved recognition rate compared with the frequency subtracted and non-subtracted cases.

• Journal of Korea Multimedia Society
• /
• v.13 no.4
• /
• pp.496-503
• /
• 2010

Speech recognition system is developed of standalone, In case of a mobile terminal using that low recognition rate represent because of limitation of memory size and audio compression. This study suggest vocabulary recognition highest performance improvement system for separate acoustic search and lexical search. Acoustic search is carry out in mobile terminal, lexical search is carry out in server processing system. feature vector of speech signal extract using GMM a phoneme execution, recognition a phoneme list transmission server using Lexical Tree Search algorithm lexical search recognition execution. System performance as a result of represent vocabulary dependence recognition rate of 98.01%, vocabulary independence recognition rate of 97.71%, represent recognition speed of 1.58 second.

• The Journal of the Acoustical Society of Korea
• /
• v.13 no.5
• /
• pp.77-82
• /
• 1994

In speech signal processing, it is very important to detect the pitch exactly. Although various methods for detecting the pitch of speech signals have been developed, it is difficult to exactly extract the pitch for wide range of speakers and various utterances. Thus we propose a new pitch detection algorithm which takes advantage of the G-peak extraction. It is a method to detect the pitch period of the voiced signals by finding MZCI (maximum zero-crossing interval) of the G-peak which is defined as cut-off bandwidth rate of LPF (low pass filter). This algorithm performs robustly with a gross error rate of 3.63% even in 0 dB SNR environement. The gross error rate for clean speech is only 0.18%. Also it is able to process all courses with high speed.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.33B no.1
• /
• pp.108-115
• /
• 1996

This paper proposes an algorithm and a model topology for the connected speech recognition using Discrete Hidden Markov Models. A proposed model uses diphone and triphone model which consider the recognition rate and recognisable vocabulary. Considering more exact inter- phoneme segmentation and execution speed of algorithm, 4 states have to exist in diphone model where the first state and the last state are keeping a steady state, the other states hold a transient state. 7 states have to exist in triphone model where 7 states are specified and improved to 3 steady states and 4 transition states. Also, the proposed speech recognition algorithm is designed to detect the inter-phoneme segmentation during the recognition processing.

• Phonetics and Speech Sciences
• /
• v.1 no.1
• /
• pp.63-67
• /
• 2009

In this paper, in order to protect music copyrights, we propose a music identification system which is scalable to the number of pieces of registered music and robust to signal-level variations of registered music. For its implementation, we define the new concepts of 'music word' and 'music phoneme' as recognition units to construct 'music acoustic models'. Then, with these concepts, we apply the HMM-based framework used in continuous speech recognition to identify the music. Each music file is transformed to a sequence of 39-dimensional vectors. This sequence of vectors is represented as ordered states with Gaussian mixtures. These ordered states are trained using Baum-Welch re-estimation method. Music files with a suspicious copyright are also transformed to a sequence of vectors. Then, the most probable music file is identified using Viterbi algorithm through the music identification network. We implemented a music identification system for 1,000 MP3 music files and tested this system with variations in terms of MP3 bit rate and music speed rate. Our proposed music identification system demonstrates robust performance to signal variations. In addition, scalability of this system is independent of the number of registered music files, since our system is based on HMM method.

• ETRI Journal
• /
• v.35 no.4
• /
• pp.697-705
• /
• 2013

Emerging trends in the area of digital very large scale integration (VLSI) signal processing can lead to a reduction in the cost of the cochlear implant. Digital signal processing algorithms are repetitively used in speech processors for filtering and encoding operations. The critical paths in these algorithms limit the performance of the speech processors. These algorithms must be transformed to accommodate processors designed to be high speed and have less area and low power. This can be realized by basing the design of the auditory filter banks for the processors on digital VLSI signal processing concepts. By applying a folding algorithm to the second-order digital gammatone filter (GTF), the number of multipliers is reduced from five to one and the number of adders is reduced from three to one, without changing the characteristics of the filter. Folded second-order filter sections are cascaded with three similar structures to realize the eighth-order digital GTF whose response is a close match to the human cochlea response. The silicon area is reduced from twenty to four multipliers and from twelve to four adders by using the folding architecture.