JPH04184398A - Voice recognizer - Google Patents

Voice recognizer

Info

Publication number
JPH04184398A
JPH04184398A JP2313547A JP31354790A JPH04184398A JP H04184398 A JPH04184398 A JP H04184398A JP 2313547 A JP2313547 A JP 2313547A JP 31354790 A JP31354790 A JP 31354790A JP H04184398 A JPH04184398 A JP H04184398A
Authority
JP
Japan
Prior art keywords
recognition
values
data memory
key
voice
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
JP2313547A
Other languages
Japanese (ja)
Other versions
JP2606443B2 (en
Inventor
Atsushi Ookumo
篤 大蜘蛛
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Panasonic Holdings Corp
Original Assignee
Matsushita Electric Industrial Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Matsushita Electric Industrial Co Ltd filed Critical Matsushita Electric Industrial Co Ltd
Priority to JP2313547A priority Critical patent/JP2606443B2/en
Publication of JPH04184398A publication Critical patent/JPH04184398A/en
Application granted granted Critical
Publication of JP2606443B2 publication Critical patent/JP2606443B2/en
Anticipated expiration legal-status Critical
Expired - Fee Related legal-status Critical Current

Links

Abstract

PURPOSE:To allow the reduction of the number of recognition standard patterns and the correction of the results of recognition and to obtain the more efficient device by adding a key input device to the voice recognizer. CONSTITUTION:Inputted voices are sent to a recognizing section 12 and the results of the recognition are sent into a data memory section 13. The values thereof are displayed on a display device 16. The value of the data memory section 13 is changed from a key operating section 16 when the numerical values exclusive of the values of the data memory section 13 are desired to be inputted. The values of he data memory section 13 are displayed on the display device 16 at every change thereof. If the many analogous words, such as numbers, exist, there is no need for having the standard patterns for all the values by adding the key input device to the voice recognizer and, therefore, the time for recognition is shortened at the time of executing the voice recognition and the decrease in the recognition rate by an increase in dictionary is suppressed. If the numerical values which are not registered in the standard patterns are desired to be obtd., the recognition is made with the values approximate thereto and the desired values are obtd. by correcting the values by key operations.

Description

【発明の詳細な説明】 産業上の利用分野 本発明C上 音声認識により得た結果をキー入力により
補正する音声認識装置に関すム 従来の技術 音声認識技術は 発展途上にある技術で、まだ完成した
技術であるとは言えないが 方式や用途によっては実用
されるものも出てきており、各種の機器に音声認識技術
が取り入れられるようになってきていも 音声認識は大体において、入力された音声と登録されて
いる各標準パターンとを比較して最も似ているものを認
識結果としていへ そのたぬ 必要とする言葉を得るに
はそのすべての標準パターンを登録しておく必要があっ
た しかし、標準パターン数を多くすると、認識に要す
る時間が多く必要となり結果を出力するのに幾らかの待
ち時間が発生してしまったり、また よく似た言葉が多
く含まれるとき地上  認識率の低下が生じてしまう。
[Detailed Description of the Invention] Industrial Application Field of the Invention C 1. Conventional technology related to a speech recognition device that corrects the result obtained by speech recognition by key input.Speech recognition technology is a developing technology and has not yet been completed. Although it cannot be said that it is a completely new technology, there are some that are put into practical use depending on the method and application, and even though voice recognition technology is being incorporated into various devices, voice recognition is generally based on the input voice. and each registered standard pattern and select the most similar one as the recognition result. However, in order to obtain the desired word, it was necessary to register all the standard patterns. If the number of standard patterns is increased, the time required for recognition will increase, resulting in some waiting time to output the results, and the ground recognition rate will decrease when many similar words are included. It will happen.

そこで、音声認識技術を実用する場合6上  ある程度
標準パターンを限定 選択する必要があった発明が解決
しようとする課題 従来の音声認識装置では 多くの数値を認識の対象とす
るような場合、登録する標準パターンの数が多くなりま
f−、Eたような言葉が多くなってしまし\ 上述した
ように待ち時間が発生してしまったり認識率の低下が起
こってしまう。
Therefore, when putting speech recognition technology into practice, it was necessary to select a standard pattern to some extent.The problem that the invention aims to solveWith conventional speech recognition devices, when many numerical values are to be recognized, it is necessary to select a standard pattern. The number of standard patterns increases, and the number of words such as f- and E increases.As mentioned above, waiting time occurs and the recognition rate decreases.

本発明は上記従来の問題点を解決するもので、音声入力
装置にキー入力機能を付加することで、より効率のよい
音声認識装置を提供することを目的とすム 課題を解決するための手段 この目的を達成するために 本発明it  音声入力を
認識する認識部と、認識部で認識された結果をキー入力
により補正するデータ補正部と、データ補正部で補正さ
れた結果データを保持する保持部と、保持部に保持され
たデータを表示する表示器とを備えた構成を有していも 作用 本発明は 上記構成により、大ざっばな数値を音声によ
り入力し、認識結果をキー入力により補正することで、
標準パターン数を必要以上に多くできないという音声認
識の欠点を補し\ 多くの数値の入力を必要とする機器
に対応する音声認識装置を実現することができも 実施例 以下、本発明の一実施例について図面を参照しながら説
明すも 図は本発明の一実施例における音声認識装置の
ブロック結線図であム 図において、 11は音声を入力するためのマイクロホ
ン、 12は入力された音声の認識処理を行う認@訊 
13はデータ記憶i  14は入力を行うキー、 15
は入力されたキーによりデータを操作するキー操作部 
16はデータを表示する表示器であム 以上のように構成された音声認識装置について、O〜9
9の数値を得るシステムで「35」という数値を入力す
る場合を例に 以下その動作を説明すも まず、認識部12に標準パターンを登録しておく必要が
あム 標準パターンには結果として得られる可能性のあ
るすべての数値を登録するのではなく、幾つ置きかの値
を登録しておく。この例ででは −船釣に使用頻度の高
いと思われム Oから10おきに10個の数値を音声認
識を行うための標準パターンとして登録すム マイクロホン11から入力された音声は 認識部12に
送られ 予め登録されている標準パターンにより認識が
行われも 認識結果はデータ記憶部13に送られ 表示
器16によりその値が表示され&  r35Jという数
値を入力する場合には「35」と発声したいところでは
ある力\ 「35」の標準パターンが認識部12に登録
されていないので「30」と発声す、4  r30Jと
正しく認識した場合、表示器に「30」が表示されも標
準パターンに登録されている以外の数値を入力したいと
きには +−キーやニューメリックキーによりキー操作
部15からデータ記憶部13の値を変更し また その
都度表示器16にはデータ記憶部13の値を表示す4 
 r35Jという数を得たい場合、認識結果の「30」
をキー操作部15をキー操作することにより「35」に
変更すム そしてこの結果を、音声認識装置に接続され
た機器(図示せず)に送も 以上のように 本実施例によれば 数字などよく似た言
葉が多く存在する場合、音声認識装置にキー入力装置を
付加することで次のような効果がえられも すべての値について標準パターンを持たなくて良いので
、音声認識を行う際に認識時間の削減ができ、また 辞
書増加による認識率の低下を抑えられも 標準パターンに登録されていない数値を得たいときはそ
れに近い値で認識を行し\ キー操作により値を修正す
ることで希望する値を得ることができも 発明の効果 以上のように本発明は 音声認識装置にキー入力装置を
付加することにより、認識標準パターン数の削減と認識
結果の補正が可能となり、より効率の良い音声認識装置
を実現することができへ
The present invention solves the above-mentioned conventional problems, and aims to provide a more efficient speech recognition device by adding a key input function to a speech input device. To achieve this object, the present invention includes a recognition unit that recognizes voice input, a data correction unit that corrects the result recognized by the recognition unit by key input, and a storage that holds result data corrected by the data correction unit. With the above structure, rough numerical values can be input by voice, and recognition results can be input by key input. By correcting
It is possible to compensate for the drawback of voice recognition in that the number of standard patterns cannot be increased more than necessary, and to realize a voice recognition device compatible with equipment that requires the input of many numerical values. An example will be explained with reference to the drawings. The figure is a block diagram of a speech recognition device according to an embodiment of the present invention. In the diagram, 11 is a microphone for inputting speech, and 12 is a recognition of input speech. Authorization for processing
13 is a data storage i; 14 is a key for input; 15
is a key operation section that operates data using input keys.
16 is a display device for displaying data.
The operation will be explained below, taking as an example the case where the numerical value ``35'' is entered in a system that obtains the numerical value 9. First, it is necessary to register a standard pattern in the recognition unit 12. Rather than registering all possible values, register every few values. In this example, the voice input from the microphone 11 is input to the recognition unit 12, where 10 values are registered every 10th from 0 as a standard pattern for voice recognition. Even if recognition is performed using the standard pattern that has been sent and registered in advance, the recognition result is sent to the data storage unit 13 and its value is displayed on the display 16. By the way, a certain force \ The standard pattern of "35" is not registered in the recognition unit 12, so "30" is uttered. 4 If it is correctly recognized as r30J, even though "30" is displayed on the display, it is registered as a standard pattern. When you want to input a value other than the one shown in the figure, change the value in the data storage section 13 from the key operation section 15 using the +- key or the numeric key, and display the value in the data storage section 13 on the display 16 each time.
If you want to get the number r35J, the recognition result "30"
is changed to "35" by operating the key operation unit 15.Then, this result is sent to a device (not shown) connected to the voice recognition device.According to this embodiment, as described above, the number When there are many similar words, such as, adding a key input device to the speech recognition device can have the following effect, but it is not necessary to have a standard pattern for all values, so when performing speech recognition, The recognition time can be reduced, and the decline in recognition rate due to the increase in dictionaries can be suppressed. However, if you want to obtain a value that is not registered in the standard pattern, perform recognition with a value close to it\\ Correct the value by key operation. By adding a key input device to the speech recognition device, the number of recognition standard patterns can be reduced and the recognition results can be corrected, making it more efficient. It is possible to realize a good speech recognition device.

【図面の簡単な説明】[Brief explanation of the drawing]

図は本発明の一実施例における音声認識装置のブロック
結線図であム 11・・・マイクロホン、 12・・・認識部 13・
・・データ記憶部 14・・・キー、 15・・・キー
操作@ 16・・・表示器
The figure is a block wiring diagram of a speech recognition device according to an embodiment of the present invention.
...Data storage unit 14...Key, 15...Key operation @ 16...Display unit

Claims (1)

【特許請求の範囲】[Claims] 音声入力を認識する認識部と、上記認識部で認識された
結果をキー入力により補正するデータ補正部と、上記デ
ータ補正部で補正された結果データを保持する保持部と
、上記保持部に保持されたデータを表示する表示器とを
備えた事を特徴とする音声認識装置。
a recognition unit that recognizes voice input; a data correction unit that corrects the result recognized by the recognition unit by key input; a holding unit that holds result data corrected by the data correction unit; and a holding unit that holds the result data corrected by the data correction unit; 1. A voice recognition device comprising: a display device for displaying the received data.
JP2313547A 1990-11-19 1990-11-19 Voice recognition device Expired - Fee Related JP2606443B2 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2313547A JP2606443B2 (en) 1990-11-19 1990-11-19 Voice recognition device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP2313547A JP2606443B2 (en) 1990-11-19 1990-11-19 Voice recognition device

Publications (2)

Publication Number Publication Date
JPH04184398A true JPH04184398A (en) 1992-07-01
JP2606443B2 JP2606443B2 (en) 1997-05-07

Family

ID=18042638

Family Applications (1)

Application Number Title Priority Date Filing Date
JP2313547A Expired - Fee Related JP2606443B2 (en) 1990-11-19 1990-11-19 Voice recognition device

Country Status (1)

Country Link
JP (1) JP2606443B2 (en)

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS56162141A (en) * 1980-05-19 1981-12-12 Nippon Telegr & Teleph Corp <Ntt> Character input device of voice input and display selection type

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS56162141A (en) * 1980-05-19 1981-12-12 Nippon Telegr & Teleph Corp <Ntt> Character input device of voice input and display selection type

Also Published As

Publication number Publication date
JP2606443B2 (en) 1997-05-07

Similar Documents

Publication Publication Date Title
DE3778579D1 (en) TRAINING OF MARKOV MODELS RELATED TO A VOICE RECOGNITION SYSTEM.
JP3426176B2 (en) Speech recognition device, method, computer system and storage medium
JPH11161464A (en) Japanese sentence preparing device
JPH04184398A (en) Voice recognizer
JP3501489B2 (en) Audio processing apparatus and method
JPH0432900A (en) Sound recognizing device
JPS6281699A (en) Forming and updating method for dictoinary for voice word processor
JPH09230889A (en) Speech recognition and response device
JP2639249B2 (en) Voice recognition device
JPH1195792A (en) Voice processing device and character inputting method
US20080262831A1 (en) Method for the Natural Language Recognition of Numbers
JPS58186836A (en) Voice input data processor
JPS60113298A (en) Voice registration system for specified speaker&#39;s voice recognition equipment
JPS5977499A (en) Voice recognition equipment
JPH0567960B2 (en)
JPS6147997A (en) Voice recognition equipment
JPH06289899A (en) Speech recognition device
JPS60159899A (en) Voice recognition equipment with learning function
JPS60200336A (en) Voice input processor
JPH07104675A (en) Recognition result display method
JPH0667688A (en) Input device
JPH05216493A (en) Operator assistance type speech recognition device
JPH0441358B2 (en)
JPS63305397A (en) Unspecified speaker voice recognition equipment
JPH0228700A (en) Voice dialing device

Legal Events

Date Code Title Description
LAPS Cancellation because of no payment of annual fees