Anti-crosstalk method, device and equipment based on multi-pickup scene
A technology for preventing crosstalk and scenes, applied in speech analysis, speech recognition, instruments, etc., can solve the problems affecting the accuracy of speech recognition processing, and achieve the effect of improving speech recognition effect, reducing the amount of calculation, and solving the problem of crosstalk.
Active Publication Date: 2020-12-29
IFLYTEK (SUZHOU) TECH CO LTD
View PDF14 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
In application environments such as meetings of various sizes, interviews, inquiries, etc., multiple sound pickup devices need to record the voice content of multiple speakers. In real scenarios, each sound pickup device is in the acquisition state, resulting in the sound pickup system not only The target sound source corresponding to a certain sound pickup device is obtained, and the interference sound source collected by other sound pickup devices is also connected in series, that is, the speech signal being processed in the sound pickup system contains a large amount of interference speech, and the sound pickup system will After this speech signal is output to the backend, it will directly affect the accuracy of speech recognition processing
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment approach
[0151] Based on the above embodiments and their preferred solutions, those skilled in the art can understand that, in actual operation, the present invention is applicable to various implementation modes, and the present invention uses the following carrier as a schematic illustration:
[0152] (1) an anti-crosstalk device, which may include:
[0153] one or more processors, memory, and one or more computer programs, wherein the one or more computer programs are stored in the memory, the one or more computer programs include instructions, when the instructions are When the device described above is executed, the device is made to perform the steps / functions of the foregoing embodiments or equivalent implementation manners.
[0154] The anti-crosstalk device may be an electronic device or a circuit device built in the above-mentioned electronic device. The aforementioned electronic device may be a PC, a server, an intelligent terminal, and the like.
[0155] In at least one p...
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 invention discloses an anti-crosstalk method, device and equipment based on a multi-pickup scene. The concept of the invention is that the balance relationship between the calculation complexity and engineering realization is fully considered, acoustic information of audio acquired by a plurality of associated sound pickup devices is selected for feature analysis, and whether crosstalk is mixed with target sound source voice or not is judged, so that serial interference voice can be suppressed at the front end. According to the invention, the crosstalk problem of the pickup devices in a specific scene can be effectively solved, and the voice recognition effect of a rear end can be significantly improved.
Description
technical field [0001] The present invention relates to the technical field of speech recognition, in particular to a method, device and equipment for preventing crosstalk based on multi-sound pickup scenarios. Background technique [0002] The premise of speech recognition is that the front-end sound pickup system collects the speech signal from the sound source, and then recognizes and transcribes the given speech signal through the back-end speech recognition algorithm. Therefore, the way to improve the accuracy of speech recognition includes at least two levels, the front-end sound pickup link and the back-end processing link. The original design intention of the present invention is to analyze, try to correct and make specific improvements to the technical defects of the front-end sound pickup link based on specific scenarios . [0003] Regarding the specific scenario, with the development of speech recognition technology, the application environment of speech recognit...
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
Login to View More IPC IPC(8): G10L15/26G10L21/02G10L21/0208
CPCG10L21/02G10L21/0208
Inventor 何超马桂林卫慧慧付建忠胡葵铭李凤霞
Owner IFLYTEK (SUZHOU) TECH CO LTD