JPS59186040A - Voice word processor - Google Patents

Abstract

PURPOSE:To improve operability of a voice word processor that recognizes monosyllables uttered continuously by detecting the end of one continuous utterance and performing kana-kanji (Japanese syllabary-Chinese character) conversion automatically by the detection signal. CONSTITUTION:Voice signals from a microphone 1 are frequency analyzed by an analyzing section 201 and sent to a recognizing section 202 together with starting signal (front end) and end signal (rear end) of utterance. The circuit 202 performs recognizing processing basing of the information and sends the result to a word processing section 3 together with the end signal. The inputted words are clause analyzed by a clause analyzing section 301. At the same time, kana- kanji conversion is performed using a dictionary section 302 basing on the end signal from the circuit 202. The result is displayed on a display section 4 through a correction and edition section 303 or printed by a printing section 5. Thus, by automatic kana-kanji conversion based on the end signal, operability can be improved.

Description

[Detailed description of the invention] The present invention is a voice word processor, 42% in number.

This invention relates to a voice word processor that inputs continuous utterances of monosyllables such as ``ha''.

In recent years, voice input has attracted attention as a new input method to replace keyboard input, and one of its application fields is voice word processing. In such voice word processors, when converting a series of single syllables into kana-kanji after voice input, as in conventional word processors that use keyboard input, you can input a series of single syllables and then convert them into kana-kanji by vocalizing, for example, "Hen-kan," or by pressing the kana-kanji conversion key. The conversion mechanism was operated by pressing a key. However, kana-kanji conversion using such means has drawbacks such as cumbersome operation and slow input speed.

The purpose of the present invention is to eliminate the drawbacks of the conventional method,
An object of the present invention is to provide a voice word processor with improved operability.

Continuously uttered monosyllables according to the invention? A voice word processor that inputs input by voice recognition, comprising an automatic kana-kanji converter groove that detects the end of one continuous utterance and automatically performs kana-kanji conversion based on the detection signal. is obtained.

In other words, according to the present invention, when a series of single syllables are manually transcribed by voice, the mechanism for converting them into kana-kanji is automatically activated, and the kanji are automatically converted into kanji, eliminating the need for any operations. Become. Therefore, operation as an audio word processor is extremely simple, and input speed is greatly improved.

The present invention will be explained in detail below with reference to the drawings. FIG. 1 is a block diagram showing an entire example of the present invention. 1st
In the figure, an audio signal from a microphone I enters an audio input section 2. In the audio input section 2, an analysis section 201 performs frequency analysis of the audio signal.

It is sent to the recognition unit 202 together with the utterance start signal 19 (starting end signal) and the utterance end 9 signal (end signal).

The recognition unit 202 performs recognition processing based on this information, and the recognition result is sent to the word processor unit 3. Additionally, a termination signal is also sent from the analysis section 201 to the word processor section 3 together with this recognition result.

In the word processor section 3, input words based on the recognition results from the voice input section 2 are subjected to phrase analysis processing in a phrase analysis processing section 301, and at the same time, are analyzed using a dictionary section 302 based on the termination signal from the voice input section 2. Kana-kanji conversion is performed. A series of sentences that have been converted into kana-kanji in the phrase analysis processing section 301 are displayed on the display section 4 or printed on the printing section 5 via the proofreading and editing processing section 303.

The voice input section 2 and word processor section 3 described here have been established in the prior art, and the point is that the terminal signal from the voice input section? This signal is sent to the word processor section, and kana-kanji conversion is automatically performed based on this signal.

The present invention has been explained above based on one embodiment, but the voice input section may be a specific speaker type or an unspecified speaker type, and the vocalization method may be completely continuous voice. Alternatively, a so-called high-speed discrete vocalization method in which syllables are uttered at intervals (pauses) of several tens of milliseconds may be used. Furthermore, if the uttered word is something other than a simple phrase, such as a number or alphabet,
It goes without saying that the termination signal should be sent to the word processor and processed so that no kana-kanji conversion is performed.

According to the present invention, an audio word processor with further improved operability can be obtained.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing one embodiment of the present invention. ■...Microphone, 2...Audio input section, 3.
...Word processor part, 4...Display part, 5...Print part\, -7/

Claims (1)

[Claims] In a speech word processor, the end of one continuous utterance is detected by recognizing continuously uttered monosyllables, and Kana-Kanji conversion is performed using the detection signal.
A voice word processor that includes automatic kana-kanji conversion @ structure.