FPGA based fast Lithuanian isolated word recognition system

Sledevič, Tomyslav; Navakauskas, Dalius

doi:10.1109/eurocon.2013.6625195

Cited by 7 publications

(9 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The first one is averaging. The 44.1 kHz sampled recordings are processed using mean filter for averaging four adjacent samples thus obtaining 11 kHz sampled speech records [14]. This allows to reduce the influence of random background noise.…”

Section: Acoustical Analysis Comparisonmentioning

confidence: 99%

“…Sakoe-Chiba [3] proposed global constraints with adjustment window of 32 frames (See Δ on Figure 3) were used in our research. This allowed us to reduce the search space by 62 % [14]. The speech recognition system prototype was implemented and all experimental testing was performed using MATLAB software.…”

Section: Discrete Fourier Transformmentioning

confidence: 99%

“…The pattern matching algorithm used in this paper is based on Dynamic Time Warping (DTW) and adopted for the implementation in FPGA based real-time system [14]. An application of the algorithm is tested on the isolated word recognition system.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Word Recognition Acceleration by Double Random Seed Matching in Perceptual Cepstrum Error Space

Serackis

Sledevič

Tamulevieius

et al. 2013

2013 European Modelling Symposium

Self Cite

View full text Add to dashboard Cite

Paper presents an algorithm for acceleration of the dynamic time warping (DTW) based isolated word recognition algorithm. The number of matching operations directly depends on the size of vocabulary. A set of perceptual cepstrum features is calculated for each word and stored in the vocabulary as a reference. Additionally all words (references) are compared between each other using DTW in order to get the reference-to-reference matches. The acceleration of pattern matching is acquired by adaptive search of the pattern reference according to the previous matching results and reference-to-reference matches. A modified word selection scenario applied for the vocabulary reduces the number of matching operations by 62-70 % in average. The reduction of matching operations allows to use DTW based speech recognition methods in real-time control applications and only need additional 13 % of vocabulary storage space.

show abstract

Section: Acoustical Analysis Comparisonmentioning

confidence: 99%

Section: Discrete Fourier Transformmentioning

confidence: 99%

See 1 more Smart Citation

Word Recognition Acceleration by Double Random Seed Matching in Perceptual Cepstrum Error Space

Serackis

Sledevič

Tamulevieius

et al. 2013

2013 European Modelling Symposium

Self Cite

View full text Add to dashboard Cite

show abstract

“…As dialect isolated word recognition systems, the comparison of recognition performance between the proposed implementation and Lithuanian system [9] is shown in Tab. II.…”

Section: A Recognition Ratementioning

confidence: 99%

“…After that, Dr. E. I. Abbas et al achieved the English digital recognition using the GMM / HMM algorithm on SOPC architecture of Altera Cyclone II family [8] . In addition, in 2013, T. Sledevic et al realized the Lithuanian real-time FPGA speech recognition system based on virtex-4 FPGA, in which DTW algorithm was used to do 600 reference templates' matching with the recognition rate up to 95% [9] .…”

Section: Introductionmentioning

confidence: 99%

The Hokkien isolated word recognition system based on FPGA

et al. 2014

2014 International Conference on Anti-Counterfeiting, Security and Identification (ASID)

View full text Add to dashboard Cite

A Hokkien isolated word recognition system has been implemented on the Spartan-6 XC6SLX45 FPGA hard-core in this paper. Firstly, an efficient method using adaptive double threshold is employed to execute the voice available detection. Secondly, the 24-th order cepstral analysis of static MFCC and differential MFCC is used to extract speech signal features. Finally Dynamic Time Warping (DTW) is utilized to find the minimum cost path between test and reference templates. The recognition system works offline and is built on medium class FPGA. The experiments on specific and non-specific speaker recognition for Hokkien isolated digits were implemented. And the analyzed results showed that the proposed system achieved a good performance in terms of recognition accuracy, calculation speed and hardware utilization.

show abstract