Speech Recognition And Its Realization Using Fix-point DSP

Posted on:2005-05-21

Degree:Master

Type:Thesis

Country:China

Candidate:X C Pang

Full Text:PDF

GTID:2168360122480352

Subject:Pattern Recognition and Intelligent Systems

Abstract/Summary:

PDF Full Text Request

In this paper, a general multi-DSP target system in ISA bus is designed and thecorresponding download software is programmed. This system can be used as hardwareworkshop for the algorithm studying for signal processing. Then the recognition methodof speech signal using isolated words is studied. The feature extraction and recognitionfor speech signal are implemented in this DSP target system, using neural network. Thefeature database is obtained by training speech signal samples from 25 isolated words.The recognition algorithm of neural network implemented in DSP target system iseffective, by testing for test speech samples. The recognition rate of isolated words canreach about 89%.

Keywords/Search Tags:

speech recognition, character extraction, neural network, DTW, algorithm, DSP, target system

PDF Full Text Request

Related items

1	Study On Chinese Character Recognition System Based On Multi-Algorithm Integration & Neural Network
2	Study Of Speech Recognition Algorithm Based On HMM And Neural Network
3	Research On The Algorithm And Realization Of Bill Character Recognition System
4	Research On Character Recognition Of License Plate Recognition System Based On Neural Networks
5	A Research Of Container Character Recognition Based On Neural Network
6	Speech Recognition Based On RBF Neural Network
7	Design And Implementation Of Robust Speech Recognition System Based On Deep Neural Network
8	Studying On Chinese Digital Speech Recognition Technology Based On Neural Network
9	Machine Printed Character Recognition System Using Feature Point Extraction And BP Neural Network Classifier
10	Handwritten English Letters Recognition Based On LVQ Neural Network