CN103455301A

CN103455301A - Voice input method

Info

Publication number: CN103455301A
Application number: CN2013103875674A
Authority: CN
Inventors: 武文
Original assignee: Individual
Current assignee: Individual
Priority date: 2013-08-31
Filing date: 2013-08-31
Publication date: 2013-12-18

Abstract

The invention relates to the field of computers, and particularly relates to a voice input method. According to the voice input method, words are divided into two parts, namely an initial pronunciation and a code, wherein the initial pronunciation represents the pronunciation of a first word in the words, and 415 Chinese pronunciations are determined according to an initial-final combination method; the code represents the code of a candidate word displayed on a screen after the initial pronunciation is given. The problems of low identification rate and low accuracy rate of an existing voice input method are thoroughly solved, so that hand operation is eliminated, a person able to speak can type, and the efficiency of the voice input method is much higher than that of the existing voice input method. An old person incapable of using a computer can immediately master the voice input method, and a young person skilled in the computer is involved in a large computer revolution.

Description

A kind of phonitic entry method

Technical field

The present invention relates to computer realm, be specifically related to a kind of phonitic entry method.

Background technology

So far, Chinese input by keyboard input methods such as developing phonetic, spelling, five, and associating input function is arranged, and the phonitic entry method of inputting by audio frequency reaches the hand-writing input method of recognizing by person's handwriting for input method development.But above several input method is Shortcomings all, phonetic and all-phonetic input method are due to a sound multiword situation of Chinese, need a plurality of candidate Chinese characters to be selected after the input Pinyin character, the Pinyin Input of phrase and whole sentence is because the difference of speech habits also needs candidate word is selected again; Though five and region code input accuracy is high but need mass memory work, the skilled people who uses is also few, and five inputs for whole sentence are supportive not high; Handwriting input is slow, and each one hand-written custom is different, also not high for the recognition correct rate of handwritten word.

Within these several years, progressively start to have occurred phonitic entry method, i.e. face typewriting, microphone input method.It can think at present easy, the most easy-to-use in the world input method, as long as you are sociable, it just can be typewrited.But also exist following shortcoming: limited by accent, identification is led not high, and wrongly written character is many, and modification is got up cumbersome, so speed is slower, and expensive price makes its scope of application that larger restriction also be arranged simultaneously, also very immature at present.The one, the phonetically similar word of Chinese is many, the difficult identification of computer; The 2nd, Chinese dialects are many, the inaccurate difficult identification of voice; The 3rd, unacquainted word can not be read out; The 4th, modification is got up pretty troublesome, still will use keyboard.After all or speech recognition and accuracy problem.

Summary of the invention

For addressing the above problem, the invention provides a kind of phonitic entry method, thoroughly solve speech recognition and accuracy problem, thereby realized high-speed and high-efficiency and the high accuracy of phonitic entry method.

The technical solution used in the present invention is: a kind of phonitic entry method is divided into two parts, first sound and code by the words input; Wherein first sound is the pronunciation of the first character of words, according to the sound method of joining, has determined 415 kinds of pronunciations of Chinese; Wherein code is after first sound sends, the code of the prepare word shown on screen.

Beneficial effect of the present invention is: thoroughly solved identification and the accuracy problem of phonitic entry method, thereby liberated both hands, realized saying just and typewriting, and efficiency has been far above existing input method.Not only solved the elderly can computer at a glance can a difficult problem, and also welcome a kind of computer great revolution to the young man who has been familiar with computer.

The accompanying drawing explanation

Fig. 1 sound method chart of joining;

Fig. 2 bo pronunciation candidate vocabulary;

Fig. 3 pu pronunciation candidate vocabulary;

Fig. 4 mo pronunciation candidate vocabulary;

Fig. 5 fo pronunciation candidate vocabulary.

Embodiment

The invention provides a kind of phonitic entry method, the words input is divided into to two parts, first sound and code; Wherein first sound is the pronunciation of the first character of words, according to the sound method of joining, has determined 415 kinds of pronunciations of Chinese as shown in Figure 1; Wherein code is after first sound sends, the code of the prepare word shown on screen as shown in Figure 2-5.

Below in conjunction with example, be specifically described:

Example 1: input " glass "

At first the user sends lead-in pronunciation " bo ", then demonstrate the commonly used words of bo pronunciation as shown in Figure 1 on display screen, then user's read the code 41, because there is not phonetically similar word in digital pronunciation, thereby can identify accurately the word " glass " of 41 codes, input complete.

Example 2: input " glucose "

At first the user sends lead-in pronunciation " pu ", then demonstrate the commonly used words of pu pronunciation as shown in Figure 1 on display screen, then user's read the code 73, because there is not phonetically similar word in digital pronunciation, thereby can identify accurately the word " glucose " of 73 codes, input complete.

Claims

1. a phonitic entry method is divided into two parts, first sound and code by the words input; Wherein first sound is the pronunciation of the first character of words, according to the sound method of joining, has determined 415 kinds of pronunciations of Chinese; Wherein code is after first sound sends, the code of the prepare word shown on screen.