Chip Design of Mel Frequency Cepstral Coefficient for Speech Recognition
碩士 === 國立暨南國際大學 === 電機工程學系 === 94 === This paper proposed the chip design of speech recognition for multimedia system. It is composed by three cores: a low-power high performance fast fourier transform (FFT) processor, a Mel-scale frequency cepstral coefficient (MFCC) circuit, and a dual-ALU digital...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | en_US |
Published: |
2006
|
Online Access: | http://ndltd.ncl.edu.tw/handle/69494964042006361607 |
id |
ndltd-TW-094NCNU0442031 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-094NCNU04420312015-10-13T10:38:05Z http://ndltd.ncl.edu.tw/handle/69494964042006361607 Chip Design of Mel Frequency Cepstral Coefficient for Speech Recognition 應用於語音辨識之梅爾倒頻譜參數超大型積體電路設計 Ying Lei 雷穎 碩士 國立暨南國際大學 電機工程學系 94 This paper proposed the chip design of speech recognition for multimedia system. It is composed by three cores: a low-power high performance fast fourier transform (FFT) processor, a Mel-scale frequency cepstral coefficient (MFCC) circuit, and a dual-ALU digital signal process (DSP) processor with dynamic time warping speech recognition algorithm. The DSP processor had been implemented by previous researcher. In this paper, we mainly proposed the FFT processor and MFCC chip. In the FFT processor, we proposed a novel register array based pipelined radix-2 structure to reduce power consumption and computation cycles. In the MFCC circuit, we adopt one pair of accumulation procedure to reduce the computation of Mel frequency bank. In addition, we also minimize the look-up table size for logarithm operations, and we use gating clock issue to reduce power consumption. The two chips are synthesized by TSMC 0.18um cell library. The die size of the FFT/IFFT processor is approximately 4.73 . And the die size of the MFCC chip is approximately 1.71 . The two chips both work at 100 MHz. Gin-Der Wu 吳俊德 2006 學位論文 ; thesis 50 en_US |
collection |
NDLTD |
language |
en_US |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 國立暨南國際大學 === 電機工程學系 === 94 === This paper proposed the chip design of speech recognition for multimedia system. It is composed by three cores: a low-power high performance fast fourier transform (FFT) processor, a Mel-scale frequency cepstral coefficient (MFCC) circuit, and a dual-ALU digital signal process (DSP) processor with dynamic time warping speech recognition algorithm. The DSP processor had been implemented by previous researcher. In this paper, we mainly proposed the FFT processor and MFCC chip. In the FFT processor, we proposed a novel register array based pipelined radix-2 structure to reduce power consumption and computation cycles. In the MFCC circuit, we adopt one pair of accumulation procedure to reduce the computation of Mel frequency bank. In addition, we also minimize the look-up table size for logarithm operations, and we use gating clock issue to reduce power consumption. The two chips are synthesized by TSMC 0.18um cell library. The die size of the FFT/IFFT processor is approximately 4.73 . And the die size of the MFCC chip is approximately 1.71 . The two chips both work at 100 MHz.
|
author2 |
Gin-Der Wu |
author_facet |
Gin-Der Wu Ying Lei 雷穎 |
author |
Ying Lei 雷穎 |
spellingShingle |
Ying Lei 雷穎 Chip Design of Mel Frequency Cepstral Coefficient for Speech Recognition |
author_sort |
Ying Lei |
title |
Chip Design of Mel Frequency Cepstral Coefficient for Speech Recognition |
title_short |
Chip Design of Mel Frequency Cepstral Coefficient for Speech Recognition |
title_full |
Chip Design of Mel Frequency Cepstral Coefficient for Speech Recognition |
title_fullStr |
Chip Design of Mel Frequency Cepstral Coefficient for Speech Recognition |
title_full_unstemmed |
Chip Design of Mel Frequency Cepstral Coefficient for Speech Recognition |
title_sort |
chip design of mel frequency cepstral coefficient for speech recognition |
publishDate |
2006 |
url |
http://ndltd.ncl.edu.tw/handle/69494964042006361607 |
work_keys_str_mv |
AT yinglei chipdesignofmelfrequencycepstralcoefficientforspeechrecognition AT léiyǐng chipdesignofmelfrequencycepstralcoefficientforspeechrecognition AT yinglei yīngyòngyúyǔyīnbiànshízhīméiěrdàopínpǔcānshùchāodàxíngjītǐdiànlùshèjì AT léiyǐng yīngyòngyúyǔyīnbiànshízhīméiěrdàopínpǔcānshùchāodàxíngjītǐdiànlùshèjì |
_version_ |
1716831509987459072 |