Speech retrieval method and system based on fast Fourier inverse transformation

A technology of inverse Fourier transform and speech, applied in speech analysis, speech recognition, digital data information retrieval, etc., can solve the problems of low accuracy and low retrieval efficiency, and achieve high-efficiency extraction, high retrieval efficiency and retrieval accuracy Effect

Inactive Publication Date: 2019-07-26
LANZHOU UNIVERSITY OF TECHNOLOGY
View PDF3 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] Based on this, it is necessary to provide a speech retrieval method and system based on inverse fast Fourier transform, in order to achieve efficient extraction of speech features with good robustness and discrimination, and to solve the problems of low retrieval efficiency and low accuracy

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Speech retrieval method and system based on fast Fourier inverse transformation
  • Speech retrieval method and system based on fast Fourier inverse transformation
  • Speech retrieval method and system based on fast Fourier inverse transformation

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0067] figure 1 It is a flow chart of the voice retrieval method based on inverse fast Fourier transform in Embodiment 1 of the present invention.

[0068] see figure 1 , the voice retrieval method based on the inverse fast Fourier transform of the embodiment, comprising:

[0069] Step S1: Obtain the voice to be queried. The voice to be queried is submitted by a mobile terminal user.

[0070] Step S2: Using Inverse Fast Fourier Transform to perform feature extraction on the speech to be queried to obtain a feature vector to be queried.

[0071] Step S3: Using the measurement matrix to reduce the dimensionality of the feature vector to be queried to obtain the feature vector to be queried after dimensionality reduction.

[0072] Specifically, a part of the Hadamard measurement matrix is ​​used to reduce the dimensionality of the feature vector to be queried.

[0073] Step S4: constructing a hash sequence to be queried according to the feature vector to be queried after dim...

Embodiment 2

[0103] The speech retrieval method based on the inverse fast Fourier transform in this embodiment includes three processes of system hash index table construction, ciphertext speech library construction, and user speech retrieval.

[0104] Step 1 System hash index table construction

[0105] 1) Perceptual hashing scheme construction

[0106] The efficient speech-aware hashing scheme proposed in this embodiment is obtained by performing inverse fast Fourier transform (IFFT) calculation on the original speech and then reducing the dimensionality of a part of the Hadamard measurement matrix. The steps of perceptual hash scheme construction are as follows:

[0107] Step 1: Preprocessing. The signal s(t)' is obtained by pre-emphasizing the speech segment s(t), which makes the frequency spectrum of the signal flatter and facilitates subsequent feature extraction.

[0108] Step 2: Framing and windowing. Divide the speech segment s(t)' into m frames of equal length and non-overlap...

Embodiment 3

[0147] Figure 4 It is a schematic structural diagram of a speech retrieval system based on inverse fast Fourier transform in Embodiment 3 of the present invention.

[0148] see Figure 4 , the speech retrieval system based on fast Fourier transform in the present embodiment, comprises:

[0149] The first voice acquisition module 401 is configured to acquire the voice to be queried.

[0150] The first feature extraction module 402 is configured to perform feature extraction on the speech to be queried by using an inverse fast Fourier transform to obtain a feature vector to be queried.

[0151] The first dimensionality reduction module 403 is configured to perform dimensionality reduction on the feature vector to be queried by using a measurement matrix to obtain a feature vector to be queried after dimensionality reduction.

[0152] The first sequence construction module 404 is configured to construct a hash sequence to be queried according to the feature vector to be queri...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a speech retrieval method and system based on fast Fourier inverse transformation. The method comprises the following steps: acquiring voice to be queried; carrying out featureextraction on the to-be-queried voice by adopting fast Fourier inverse transformation to obtain a to-be-queried feature vector; carrying out dimension reduction on the to-be-queried feature vector byadopting the measurement matrix to obtain a dimension-reduced to-be-queried feature vector; constructing a to-be-queried hash sequence according to the to-be-queried feature vector subjected to dimension reduction; matching the to-be-queried Hash sequence with a system Hash index table, and determining a matched Hash sequence; determining a retrieval voice file according to the matched hash sequence and the ciphertext voice library; and storing the system hash index table and the ciphertext voice library in the cloud server. According to the method, the fast Fourier inverse transformation iscombined with the measurement matrix, so that the speech features with good robustness and distinction can be efficiently extracted, and the retrieval efficiency and the retrieval accuracy are improved.

Description

technical field [0001] The invention relates to the technical field of voice retrieval, in particular to a voice retrieval method and system based on inverse fast Fourier transform. Background technique [0002] With the rapid development of Internet technology, massive multimedia information emerges as the times require, especially unstructured data such as voice, which is growing exponentially. How to retrieve data from massive voice information, and how to efficiently and accurately retrieve data under the premise of ensuring the security of massive voice information are important challenges in the field of voice retrieval. [0003] At present, there are many research results in speech retrieval technology. The technology is mainly divided into: text-based or keyword retrieval, content-based retrieval. Content-based speech retrieval can be divided into: feature matching, deep learning, sorting retrieval, etc. Feature extraction is an important step in speech retrieval....

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06F16/683G10L15/02G10L15/08G10L25/18
CPCG06F16/683G10L15/02G10L15/08G10L25/18G10L2015/081
Inventor 张秋余葛子贤胡颖杰张其文李昱洲赵雪娇白建许福久
Owner LANZHOU UNIVERSITY OF TECHNOLOGY
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap