Deprecated: The each() function is deprecated. This message will be suppressed on further calls in /home/zhenxiangba/zhenxiangba.com/public_html/phproxy-improved-master/index.php on line 456
JPH0437995B2 - - Google Patents
[go: Go Back, main page]

JPH0437995B2 - - Google Patents

Info

Publication number
JPH0437995B2
JPH0437995B2 JP57157236A JP15723682A JPH0437995B2 JP H0437995 B2 JPH0437995 B2 JP H0437995B2 JP 57157236 A JP57157236 A JP 57157236A JP 15723682 A JP15723682 A JP 15723682A JP H0437995 B2 JPH0437995 B2 JP H0437995B2
Authority
JP
Japan
Prior art keywords
word
syllables
words
dictionary
registration
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.)
Expired
Application number
JP57157236A
Other languages
Japanese (ja)
Other versions
JPS5946699A (en
Inventor
Akihiro Kimura
Juji Kijima
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.)
Fujitsu Ltd
Original Assignee
Fujitsu 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 Fujitsu Ltd filed Critical Fujitsu Ltd
Priority to JP57157236A priority Critical patent/JPS5946699A/en
Publication of JPS5946699A publication Critical patent/JPS5946699A/en
Publication of JPH0437995B2 publication Critical patent/JPH0437995B2/ja
Granted legal-status Critical Current

Links

Description

【発明の詳細な説明】[Detailed description of the invention]

(1) 発明の技術分野 本発明は単語登録の際に音節数の差により単語
の認識率が異なることを考慮して総合的に認識率
を向上させる単語音声登録装置に関するものであ
る。 (2) 従来技術と問題点 従来の単語音声認識装置としては、各単語を1
回登録する方式と、全単語を複数回登録する方式
があるが、前者はとくに音節数の少ない単語に関
して十分な認識率が得られないという欠点があ
り、後者はその点は改善されるが登録回数が多く
手順と時間がかかり過ぎるという欠点があつた。 (3) 発明の目的 本発明の目的は単語登録の際に音節数の差によ
り単語の認識率が異なることを考慮して総合的に
認識率を向上させる単語音声登録装置を提供する
ことである。 (4) 発明の構成 前記目的を達成するため、本発明は、単語音声
を登録して認識を行なう単語音声認識装置におけ
る単語音声登録装置であつて、音節数の少ない単
語の登録を複数回行ない複数個の辞書内容または
その平均の辞書内容を作成する手段を設けたこと
を特徴とするものである。 (5) 発明の実施例 第1図、第2図は本発明の原理説明図である。
第1図は単語音声の音節数と認識率の関係を示す
図面である。すなわち、一般に単語音声の音節数
が少ないと認識率が低い。これに対し辞書の登録
回数を1回、2回、3回と増加した辞書内容を用
意すると同図,,のように認識率は向上す
る。また音節数の多い単語音声は登録回数を増し
ても認識率の向上は見られない。 そこで、音節数の少ない単語の登録のみを複数
回行なうようにし、音節数の多い単語は1回登録
で済ませるようにすればよい。 第2図はこの方式の1例を示すものである。す
なわち、1音節と2音節の単語を3回登録、3
音節と4音節の単語を2回登録、5音節以上の
単語は1回登録すれば、全単語に対し登録回数を
余り増加させることなく高い認識率を保持するこ
とができる。 第3図は本発明の実施例の構成説明図である。
同図において、単語名・音節数リストメモリ10
の各アドレスには単語名コードとその単語の音節
数が格納されており、後述の単語名音節数リスト
メモリアドレスカウンタ20で指定された内容
は、登録指示語レジスタ11および音節数レジス
タ12にそれぞれ読出される。登録指示語レジス
タ11の内容はデイスプレイ装置13に表示され
る。音節数レジスタ12の値からは音節数登録回
数変換ROM14を用いて、その単語の登録回数
が求められ、その値は登録回数カウンタ15に格
納される。下表に音節数登録回数変換ROM14
の内容の1例を示す。
(1) Technical Field of the Invention The present invention relates to a word speech registration device that comprehensively improves the recognition rate by taking into consideration that the recognition rate of words varies depending on the difference in the number of syllables when registering words. (2) Conventional technology and problems Conventional word speech recognition devices
There are two methods: one in which all words are registered multiple times, and one in which all words are registered multiple times, but the former has the disadvantage that a sufficient recognition rate cannot be obtained, especially for words with a small number of syllables, while the latter improves this point, but The drawback was that it required many steps and took too much time. (3) Purpose of the Invention The purpose of the present invention is to provide a word speech registration device that comprehensively improves the recognition rate by taking into account that the recognition rate of words varies depending on the number of syllables when registering words. . (4) Structure of the Invention In order to achieve the above object, the present invention is a word speech registration device in a word speech recognition device that registers word sounds and performs recognition, which registers words with a small number of syllables multiple times. The present invention is characterized by providing means for creating a plurality of dictionary contents or the average dictionary contents. (5) Embodiments of the invention FIGS. 1 and 2 are diagrams explaining the principle of the present invention.
FIG. 1 is a diagram showing the relationship between the number of syllables of word speech and the recognition rate. That is, in general, the recognition rate is low when the number of syllables in a word is small. On the other hand, if dictionary contents are prepared in which the number of dictionary registrations is increased by one, two, or three times, the recognition rate improves as shown in the figure. Furthermore, for word sounds with a large number of syllables, the recognition rate does not improve even if the number of registrations is increased. Therefore, only words with a small number of syllables may be registered multiple times, and words with a large number of syllables may be registered only once. FIG. 2 shows an example of this method. In other words, 1-syllable and 2-syllable words are registered 3 times, 3
By registering syllables and words with 4 syllables twice and registering words with 5 or more syllables once, a high recognition rate can be maintained without significantly increasing the number of registrations for all words. FIG. 3 is an explanatory diagram of the configuration of an embodiment of the present invention.
In the same figure, word name/syllable number list memory 10
A word name code and the number of syllables of the word are stored in each address, and the contents specified by the word name syllable number list memory address counter 20 (described later) are stored in the registration instruction word register 11 and the syllable number register 12, respectively. Read out. The contents of the registration instruction word register 11 are displayed on the display device 13. From the value of the syllable number register 12, the number of registrations of the word is determined using the syllable number registration number conversion ROM 14, and the value is stored in the registration number counter 15. The table below shows the number of syllables registered.Conversion ROM14
An example of the content is shown below.

【表】 一方、デイスプレイ装置13に表示された登録
指示語に従つて発声された音声はマイクロホーン
16によつて音声分析装置17に送られ分析が行
なわれ、特徴抽出装置18で辞書パターンに変換
される。辞書パターンは登録指示語レジスタ11
の内容とともに、辞書メモリアドレスカウンタ2
1で指定される辞書メモリ22に格納される。辞
書メモリアドレスカウンタ21は音声登録パルス
でインクリメントされ、登録回数カウンタ15は
音声登録パルスでデイクリメントされる。次のゼ
ロ検出部19は登録回数カウンタ15の内容がゼ
ロになつた時にパルスを出力し、単語名音節数リ
ストメモリアドレスカウンタ20をインクリメン
トし、登録指示語を次の単語にする。以上の動作
を繰返すことによつて単語登録を行なう。 本実施例によれば、音節数登録回数変換に
ROMを用いているので、音節数と登録回数の対
応関係を、単語名音節数リストを変更することな
く、ROMを取換えるだけで自由に変更できると
いう利点がある。 実施例では音節数の少ない単語の登録を複数回
行なうものであるが、これら複数の登録単語の内
容の平均を求めて辞書に格納することにより、同
様の効果を少ない辞書容量で実現することができ
る。 (6) 発明の効果 以上説明したように、本発明によれば、単語音
声登録の際音節数の少ない単語の登録回数を多く
とるようにすることにより、総合的に比較的少な
い登録回数で高い認識率を保持することができ
る。
[Table] On the other hand, the voice uttered according to the registered instruction word displayed on the display device 13 is sent to the voice analysis device 17 via the microphone 16 for analysis, and then converted into a dictionary pattern by the feature extraction device 18. be done. The dictionary pattern is registered instruction word register 11
along with the contents of dictionary memory address counter 2
It is stored in the dictionary memory 22 designated by 1. The dictionary memory address counter 21 is incremented by the voice registration pulse, and the registration number counter 15 is decremented by the voice registration pulse. The next zero detection unit 19 outputs a pulse when the content of the registration number counter 15 becomes zero, increments the word name syllable number list memory address counter 20, and sets the registration instruction word to the next word. Word registration is performed by repeating the above operations. According to this embodiment, the conversion of the number of syllables registered
Since a ROM is used, there is an advantage that the correspondence between the number of syllables and the number of registrations can be changed freely by simply replacing the ROM without changing the word name syllable number list. In the embodiment, words with a small number of syllables are registered multiple times, but by calculating the average of the contents of these multiple registered words and storing it in the dictionary, the same effect can be achieved with less dictionary capacity. can. (6) Effects of the Invention As explained above, according to the present invention, by increasing the number of registrations for words with a small number of syllables when registering word sounds, a high number of registrations can be achieved with a relatively small overall number of registrations. Recognition rate can be maintained.

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

第1図、第2図は本発明の原理説明図、第3図
は本発明の実施例の構成説明図であり、図中、1
0は単語名音節数リストメモリ、11は登録指示
語レジスタ、12は音節数レジスタ、13はデイ
スプレイ装置、14は音節数登録回数変換
ROM、15は登録回数カウンタ、16はマイク
ロホーン、17は音声分析装置、18は特徴抽出
装置、19はゼロ検出部、20は単語名音節数リ
ストメモリアドレスカウンタ、21は辞書メモリ
アドレスカウンタ、22は辞書メモリを示す。
1 and 2 are diagrams for explaining the principle of the present invention, and FIG. 3 is a diagram for explaining the configuration of an embodiment of the present invention.
0 is word name syllable number list memory, 11 is registration instruction word register, 12 is syllable number register, 13 is display device, 14 is syllable number registration number conversion
ROM, 15 is a registration number counter, 16 is a microphone, 17 is a speech analysis device, 18 is a feature extraction device, 19 is a zero detection unit, 20 is a word name syllable number list memory address counter, 21 is a dictionary memory address counter, 22 indicates dictionary memory.

Claims (1)

【特許請求の範囲】[Claims] 1 単語音声を登録して認識を行なう単語音声認
識装置における単語音声登録装置であつて、音節
数の少ない単語の登録を複数回行ない複数個の辞
書内容またはその平均の辞書内容を作成する手段
を設けたことを特徴とする単語音声登録装置。
1. A word speech registration device in a word speech recognition device that registers word speech and performs recognition, which has means for registering words with a small number of syllables multiple times to create a plurality of dictionary contents or an average dictionary content. A word voice registration device characterized by:
JP57157236A 1982-09-09 1982-09-09 Word voice registration system Granted JPS5946699A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP57157236A JPS5946699A (en) 1982-09-09 1982-09-09 Word voice registration system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP57157236A JPS5946699A (en) 1982-09-09 1982-09-09 Word voice registration system

Publications (2)

Publication Number Publication Date
JPS5946699A JPS5946699A (en) 1984-03-16
JPH0437995B2 true JPH0437995B2 (en) 1992-06-23

Family

ID=15645214

Family Applications (1)

Application Number Title Priority Date Filing Date
JP57157236A Granted JPS5946699A (en) 1982-09-09 1982-09-09 Word voice registration system

Country Status (1)

Country Link
JP (1) JPS5946699A (en)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS61249182A (en) * 1985-04-27 1986-11-06 Toshiba Corp Pattern recognizing and learning device
JP2647063B2 (en) * 1995-09-25 1997-08-27 井関農機株式会社 Voice recognition device
JP2002108389A (en) * 2000-09-29 2002-04-10 Matsushita Electric Ind Co Ltd Method and apparatus for searching and extracting personal names by voice and on-vehicle navigation device

Also Published As

Publication number Publication date
JPS5946699A (en) 1984-03-16

Similar Documents

Publication Publication Date Title
US5353377A (en) Speech recognition system having an interface to a host computer bus for direct access to the host memory
US3632887A (en) Printed data to speech synthesizer using phoneme-pair comparison
CN108597493B (en) Audio exchange method and audio exchange system of language semantics
JPH0437995B2 (en)
JPH05100693A (en) Computer-system for speech recognition
JPS5663667A (en) Voice interpreter
JPS61239358A (en) Documentation system by voice input
JPS61236598A (en) Word voice registration system
JPS58134371A (en) Japanese word input device
JPS6232798B2 (en)
JPS60115993A (en) Monosyllabic voice recognition equipment
JPS60164800A (en) Voice recognition equipment
JPS6045298A (en) Word voice recognition equipment
Hall Jr The Pronunciation of Italian: A Practical Introduction
JPH07160685A (en) Text-to-speech device
JPS6153699A (en) voice recognition device
JPS60140396A (en) Voice guidance system
JPH0236467A (en) Word reading system
JPS63192155A (en) document display device
JPS5878199A (en) Voice output unit
JPS63118794A (en) Voice synthesizer
JPS59194766U (en) character learning machine
JPS58116595A (en) word speech recognizer
JPS63249899A (en) Voice recognition word specifying system
JPS6155760U (en)