Signal front-end processing method and device for speech recognition
A speech signal and speech recognition technology, applied in speech analysis, instruments, etc., can solve the problems of divergence of the adaptive filter of the echo cancellation module, failure to work normally, and difficulties in the speech recognition system, so as to improve the robustness and ensure reliability and stability. the effect of running
Active Publication Date: 2017-07-25
BAIDU ONLINE NETWORK TECH (BEIJIBG) CO LTD
View PDF6 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
In a real sound field environment, especially in far-field pickup, the voice received by the microphone usually has different degrees of echo, which brings great difficulties to the existing speech recognition system
[0003] In order to improve the performance of speech recognition, an echo cancellation module is added to the front end of the speech recognition system to cancel the echo interference signal, but when the microphone and the speaker circuit are not synchronized, the adaptive filter in the echo cancellation module will diverge and cannot work normally
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 Construction
[0023] Embodiments of the present application are described in detail below, examples of which are shown in the drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the figures are exemplary, and are intended to explain the present application, and should not be construed as limiting the present application.
[0024] The signal front-end processing method and device for speech recognition according to the embodiments of the present application will be described below with reference to the accompanying drawings.
[0025] figure 1 It is a flowchart of a signal front-end processing method for speech recognition according to an embodiment of the present application.
[0026] Such as figure 1 As shown, the signal front-end processing method for speech recognition includes:
[0027] Step 101, if it is detected that there is a voice s...
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 application provides a signal front-end processing method used for voice recognition and a device thereof. The method comprises the steps that echo path time delay is determined according to first time at which a loudspeaker plays signals and second time at which a microphone receives etch signals if voice signals exist through detection; and time delay compensation is performed on the echo signals according to the echo path time delay and processing frame length of a self-adaptive filter, and the echo signals are tracked for elimination. According to the signal front-end processing method used for voice recognition and the device thereof, real-time tracking of time delay difference between the loudspeaker and the microphone can be realized, reliable and stable operation of the self-adaptive filter can be guaranteed, and robustness of voice system recognition can be enhanced.
Description
technical field [0001] The present application relates to the technical field of speech recognition processing, in particular to a signal front-end processing method and device for speech recognition. Background technique [0002] With the development of speech recognition technology, the application fields of speech recognition system are becoming wider and wider, such as: vehicle speech recognition system, far-field speech recognition system, speech input method system and smart home system. In a real sound field environment, especially in a far-field pickup, the speech received by the microphone usually has different degrees of echo, which brings great difficulties to the existing speech recognition system. [0003] In order to improve the performance of speech recognition, an echo cancellation module is added to the front end of the speech recognition system to cancel the echo interference signal, but when the microphone and the speaker circuit are not synchronized, the ...
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 Patent Type & Authority Patents(China)
IPC IPC(8): G10L21/0216
Inventor 崔玮玮徐杨飞魏建强
Owner BAIDU ONLINE NETWORK TECH (BEIJIBG) CO LTD