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

Speech input method and system

A voice input and voice fragment technology, applied in voice input/output, voice analysis, voice recognition and other directions, can solve problems such as unfriendly and cumbersome users

Active Publication Date: 2018-07-31
SHANGHAI GUOKE ELECTRONICS
View PDF12 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Therefore, this confirmation process is usually cumbersome and not friendly enough for users

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
  • Speech input method and system
  • Speech input method and system
  • Speech input method and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0048] Such as Figure 3~5 As shown, the present invention provides a voice input method, including:

[0049] Step S11: While recording, the input voice is continuously divided into voice fragments and the text of each voice fragment is generated. Specifically, the present invention can automatically split the voice recognition result and perform the segment return for the user to confirm again. The server continuously divides the input voice into voice fragments and generates the text of each voice fragment, and continuously divides the input voice into voice fragments through the voice endpoint detection algorithm. The voice endpoint detection accurately determines the voice from a signal containing voice The starting point and the ending point distinguish between voice and non-voice signals. Voice endpoint detection is an important aspect of voice processing technology. For example, when the user continuously inputs voice, the cloud server can use the endpoint detection algori...

Embodiment 2

[0061] Such as Image 6 with Figure 7 As shown, the present invention provides another voice input method. The difference between this embodiment and the embodiment is that it adds a step of performing noise monitoring on the recording environment to obtain the signal-to-noise ratio during recording, and the candidates can be adjusted according to different signal-to-noise ratios. The number of results, and prompts the user in the case of strong noise that is not suitable for voice input. This example can specifically include:

[0062] Step S21: Perform noise monitoring on the recording environment during recording to obtain the signal-to-noise ratio. Specifically, this step can automatically detect the signal-to-noise ratio of the input voice and feed it back on the interactive interface, which can be used in the case of strong noise that is not suitable for voice input The user is reminded that the number of candidate results can also be adjusted according to different signal-t...

Embodiment 3

[0077] Such as Figure 8 As shown, the present invention also provides another voice input system, including a segmentation module 41, a correction module 42, and a noise monitoring unit 43.

[0078] The segmentation module 41 is used to continuously segment the input voice into voice segments and generate the text of each voice segment while recording. Specifically, the segmentation module 41 is located on a cloud server, and the segmentation module 41 uses the voice The endpoint detection algorithm continuously divides the input voice into voice segments. This module can automatically split the voice recognition results and return them in segments for the user to confirm again.

[0079] The correction module 42 is used to display the text of each voice segment in turn, and correct the text of each voice segment in turn according to the user's selection. Specifically, this module can enable the user to modify and confirm the returned text while recording. However, in the interacti...

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 present invention relates to a voice input method and system. The method comprises: continuously dividing the input voice into voice segments and generating the text of each voice segment while recording; and sequentially displaying the text of each voice segment, according to the user The selection of , in turn, corrects the text of each speech segment. The invention can automatically segment the voice recognition result and return it in segments for the user's second confirmation, and the user can modify and confirm the returned text while recording.

Description

Technical field [0001] The invention belongs to the field of speech recognition, and particularly relates to a speech input method and system. Background technique [0002] With the advancement of voice recognition technology and the rise of cloud computing, it has become a trend to use voice input on mobile terminals and perform voice-to-text transcription through cloud servers and return the text to the mobile terminal. Due to the size limitation of mobile terminals, the convenience of text input directly through a physical or virtual keyboard is always unsatisfactory. It is foreseeable that voice input will replace key input in more and more places. [0003] However, the current situation that the accuracy of speech recognition is difficult to reach 100% has hindered the process of speech input completely replacing key input. In fact, due to the complexity of real pronunciation under various conditions in life, the accuracy of speech recognition can never reach 100%, especially...

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 Patents(China)
IPC IPC(8): G10L15/26G06F3/16
Inventor 李曜许东星
Owner SHANGHAI GUOKE ELECTRONICS