Method for adaptively improving speech recognition rate by means of gain
A speech recognition and self-adaptive technology, applied in speech recognition, speech analysis, instruments, etc., can solve problems such as unstoppable channels, easy saturation of channels, and distortion of recording data, so as to improve recognition rate and task completion rate, suppress channel saturation, The effect of reducing recording gain
Active Publication Date: 2006-07-12
INST OF AUTOMATION CHINESE ACAD OF SCI
View PDF0 Cites 20 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0003] However, various speech recognition methods and products do not perform well in noisy environments. There are mainly the following problems: First, the false triggering problem in noisy environments, it is easy to mistake the noisy background sound for speaking
The second is the recognition rate problem in the noise environment: the signal-to-noise ratio is relatively low, resulting in a low recognition rate
The third is the problem of judging the end of speech in a noisy environment: it is easy to mistake the noisy background sound for talking all the time, so it cannot end for a long time
The fourth is the problem of channel saturation in noisy environments: when the external environment is noisy, the channel is easily saturated, which will distort the recording data and affect speech recognition
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
[0028] figure 1 It is the comparison data (percentage data) of task completion rate before and after using this method.
[0029] Record a batch of data comparison results in front of the TV, on the city railway, and on the road. It can effectively improve the speech recognition rate in various noise environments.
[0030] figure 2 The method for improving the speech recognition rate by using gain adaptation, the steps are as follows:
[0031] Step S1, evaluating background noise;
[0032] Step S2, adjusting the recording gain according to the type of background noise evaluated in step S1;
[0033] In step S3, endpoint detection and voice recognition are performed on the basis of steps S1 and S2.
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 relates to a method for improving speech recognition rate with gain self-adapting. The invention is characterized in that: by evaluating noise, adjusting recording gain and adjusting port detection parameters to improve speech rate. The inventive method comprises steps of: 1, evaluating background noise; 2, adjusting recording gain according to the evaluated background noise in step 1; taking port detection and speech recognition on the basis of step 1 and step 2.
Description
technical field [0001] The invention relates to the field of automatic speech recognition, in particular to a method for improving speech recognition rate by using gain self-adaptation. Background technique [0002] Speech recognition technology has been greatly developed in recent years, and it has been widely used in embedded fields such as mobile phones. Among them, the voice dialer has gradually become one of the necessary functions of high-end smart phones. It is convenient and fast to use voice dialing, and there is no need to memorize cumbersome phone numbers, and it is not necessary to look up contacts page by page on the screen, which greatly improves people's communication efficiency. [0003] However, various speech recognition methods and products do not perform well in noisy environments. There are mainly problems in the following aspects: First, false triggers in noisy environments. It is easy to mistake noisy background sounds for speaking. The second is 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 IPC IPC(8): G10L21/02G10L15/20H04M9/08G10L21/0216G10L25/87
Inventor 徐波谢传泉张东泉普剑涛张亮张建
Owner INST OF AUTOMATION CHINESE ACAD OF SCI
PatSnap Eureka turns technology decisions into work you can execute. Powered by our Innovation Knowledge Graph, it runs expert workflows across engineering, life sciences, materials and intellectual property. Get your review-ready output in minutes.
