Unlock instant, AI-driven research and patent intelligence for your innovation.

Voiceprint data transmission and recognition method, recognition device, and storage medium

A transmission method and voiceprint technology, which is applied in voiceprint data recognition devices, readable storage media, voiceprint data transmission, and voiceprint data recognition fields, and can solve problems such as time-consuming transmission and string length overflow

Active Publication Date: 2019-01-04
ALLWINNER TECH CO LTD
View PDF16 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In addition to the existence of RS encoding, the actual output string length may overflow to a certain extent, and the transmission will take longer

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
  • Voiceprint data transmission and recognition method, recognition device, and storage medium
  • Voiceprint data transmission and recognition method, recognition device, and storage medium
  • Voiceprint data transmission and recognition method, recognition device, and storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0048] refer to figure 1 , first execute step S1, receive the voiceprint data (PCM data), cut the PCM data, and divide it into N equal parts in the time domain into {PCM i}| 1≤i≤N , in this embodiment, the data is divided into three parts, that is, the voiceprint data is divided into a plurality of segmented time domain data in the time domain, and the first segmented time domain data PCM1, the second segmented time domain data PCM2 and the second segmented time domain data are obtained. Segmented time domain data PCM3.

[0049] Then step S2 is executed, and the W frequency domain range is also divided into N equal parts {W i}| 1≤i≤N , in this embodiment, the frequency domain range is divided into three parts, taking frequency response W=(0, 22Khz), N=3 as an example.

[0050] W is split into W 1 = (0Khz, 7.5Khz), W 2 =(7.5khz,15khz),W 3 =(15khz,22khz).

[0051] from figure 2 The frequency-semi-symbol index mapping table in i The frequency point information I i Con...

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 provides a voiceprint data transmission and recognition method, a recognition device and a storage medium. The transmission method comprises the steps: dividing voiceprint data into a plurality of pieces of segmented time domain data in a time domain; extracting each piece of segment time domain data in a target domain; and transmitting each piece of segmented frequency domain data in different transmission frequency domains. Since the frequency point information of the voiceprint data is mainly concentrated in the target frequency domain, only the data in the target frequency domain needs to be transmitted, and the data in the non-target frequency domain is not transmitted, and the frequency domain in the non-target frequency domain is used for the transmission of other segmented frequency domain data. The voiceprint data is firstly divided into the plurality of pieces of segmented time domain data in the time domain, and then the segmented frequency domain data of the target frequency domain is extracted, and finally the segmented frequency domain data is transmitted in different transmission frequency domains, and is reused at each frequency band, thereby dividinga character string into multiple segments, outputting the segments in parallel in the full frequency domain, and reducing the data transmission time.

Description

technical field [0001] The invention relates to the field of voiceprint recognition, in particular to a voiceprint data transmission method, a voiceprint data recognition method, a voiceprint data recognition device and a readable storage medium. Background technique [0002] The frequency range of sound that can be recognized by the human ear is approximately 20 Hz to 20000 Hz. The low frequency part is 20Hz to 200Hz, the mid frequency part is 500Hz to 2kHz, and the high frequency part is 2kHz to 20kHz. [0003] Noise refers to the continuous irregular sound caused by objects colliding, compressing, etc. spontaneously or interactively in real life. Noise occurs in every frequency band, which is roughly divided into low-frequency and high-frequency noise. Due to the wavelength, low-frequency noise has a long wavelength, which exists in a large number in reality, and has strong penetration. There will be a lot of irregular energy in the low frequency part. On the contrary,...

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): G10L17/00G10L21/0224G10L21/0232G10L19/16
CPCG10L17/00G10L19/167G10L21/0224G10L21/0232
Inventor 程衎
Owner ALLWINNER TECH CO LTD