Speaker rapid identification method and system based on growing and clustering algorithm of models
What is Al technical title?
Al technical title is built by PatSnap Al team. It summarizes the technical point description of the patent document.
A recognition method and model recognition technology, applied in speech analysis, instruments, etc., can solve problems such as long matching time and poor real-time performance
Inactive Publication Date: 2015-11-25
GUANGDONG UNIVERSITY OF FOREIGN STUDIES
View PDF6 Cites 15 Cited by
Summary
Abstract
Description
Claims
Application Information
AI Technical Summary
This helps you quickly interpret patents by identifying the three key elements:
Problems solved by technology
Method used
Benefits of technology
Problems solved by technology
[0004] The primary purpose of the present invention is to overcome the defects of long matching time and poor real-time performance described in the above-
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
Click on the blue label to locate the original text in one second.
Reading with bidirectional positioning of images and text.
Smart Image
Examples
Experimental program
Comparison scheme
Effect test
Example Embodiment
[0069] Example 1
[0070] Such as figure 1 As shown, a rapid speaker recognition method based on model growth clustering includes model training and model recognition;
[0071] Model training includes the following steps:
[0072] S1: Collect the voiceprint signals of multiple people including the speaker, that is, voice signals;
[0073] S2: Perform pre-processing and noise reduction processing on each voiceprint signal, and the pre-processing process includes pre-emphasis, framing, windowing and endpoint detection in turn;
[0074] In the specific implementation process, in step S2, preprocessing each voiceprint signal specifically includes the following steps:
[0075] S2.1: Pre-emphasis, in the pre-emphasis process, the voiceprint signal is moved to the appropriate frequency band through the filter,
[0076] The transfer function is: H(z) = 1-0.9375z -1 ,
[0077] The signal obtained is: S ~ ( n ) = S ( n ) - 0.9375 S ( n - 1 ) ;
[007...
Example Embodiment
[0113] Example 2
[0114] Such as image 3 As shown, a rapid speaker recognition system based on model growth clustering includes: a client, a network connection module, and a server. The client and the server are connected through the network connection module;
[0115] The client includes:
[0116] Voiceprint acquisition module: used to collect voiceprint signals of multiple people including the speaker and output to the preprocessing module;
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
Login to view more
Abstract
The present invention discloses a speaker rapid identification method and a system based on the growing and clustering algorithm of models. The method comprises the processes of model training and model identification. The model training process comprises the steps of acquiring voiceprint signals from multiple persons including a speaker, pre-treating all the voice-print signals and extracting voiceprint characteristic parameters to form a plurality of models, and conducting the adaptive classification for all models based on the growing and clustering algorithm of models. The model identification process comprises the steps of acquiring voice signals from a speaker, pre-treating the voice signals, extracting voiceprint characteristic parameters, calculating the characteristic parameters of to-be-identified voice signals and the likelihoods of all model types, selecting a model type for the to-be-identified voice signal based on the maximum likelihood principle, calculating the likelihood scores of all models in the above selected model type, and adopting a model of the highest likelihood score as an identification result. According to the technical scheme of the invention, the operation of matching the to-be-identified voice characteristics with all models is not required, so that the method is short in matching period and good in real-time performance. The method can be well adapted to large-scale model bases.
Description
Technical field [0001] The present invention relates to the field of voiceprint recognition, and more specifically, to a method and system for rapid speaker recognition based on model growth clustering. Background technique [0002] In the embedded operating system to realize the identification of the speaker's identity through voice, it is usually necessary to preprocess the input voiceprint, transmit the data to the server, and then generate the voiceprint model, match the model, and finally output and display the result. Among them, the voiceprint model refers to the Gaussian Mixture Model (GMM), which uses the EM algorithm for training. Generally, λ=(ω,μ,Σ) triples can be used to concisely express a Gaussian mixture model. The Gaussian mixture model uses a weighted combination of multiple Gaussian models to describe a speaker's speech model, and uses the local expected maximum algorithm EM to continuously update system parameters, thereby obtaining the approximate mathematic...
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
Application Date:The date an application was filed.
Publication Date:The date a patent or application was officially published.
First Publication Date:The earliest publication date of a patent with the same application number.
Issue Date:Publication date of the patent grant document.
PCT Entry Date:The Entry date of PCT National Phase.
Estimated Expiry Date:The statutory expiry date of a patent right according to the Patent Law, and it is the longest term of protection that the patent right can achieve without the termination of the patent right due to other reasons(Term extension factor has been taken into account ).
Invalid Date:Actual expiry date is based on effective date or publication date of legal transaction data of invalid patent.