JPH01197795A - Voice recognizing device - Google Patents
Voice recognizing deviceInfo
- Publication number
- JPH01197795A JPH01197795A JP63022446A JP2244688A JPH01197795A JP H01197795 A JPH01197795 A JP H01197795A JP 63022446 A JP63022446 A JP 63022446A JP 2244688 A JP2244688 A JP 2244688A JP H01197795 A JPH01197795 A JP H01197795A
- Authority
- JP
- Japan
- Prior art keywords
- voice
- response
- recognition
- registered
- dictionary
- 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.)
- Pending
Links
- 238000010586 diagram Methods 0.000 description 3
- 206010011224 Cough Diseases 0.000 description 1
Abstract
Description
【発明の詳細な説明】
瑳４光訪
本発明は、音声認識装置、より詳細には、音声ダイヤリ
ング装置等におけるアンサーバック用応答装置に関する
。DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a voice recognition device, and more particularly to a response device for answerback in a voice dialing device or the like.
従】」Ｕｌ
音声認識の認識結果を音声でアンサーバックするために
、その応答音声を登録する必要がある。In order to answer back the recognition result of voice recognition by voice, it is necessary to register the response voice.
通常、音声認識では辞書作成のため、認識させたい単語
を数回ずつ登録し、応答用音声はさらに別に登録してい
た。すなわち、音声認識を用いたシステムでは、認識率
が１００％でない限り、認識結果を話者に何らかの形で
知らせる必要がある。Normally, in speech recognition, in order to create a dictionary, words to be recognized are registered several times at a time, and response voices are registered separately. That is, in a system using speech recognition, unless the recognition rate is 100%, it is necessary to inform the speaker of the recognition result in some form.
そのためには、応答用の音声を一度登録する必要がある
。そのシステムがあらかじめ認識する単語を固定してい
る場合はその単語のデータをあらかじめ持っていればよ
い、しかし、ユーザが認識する単語を登録するような場
合、応答用音声も登録しなくてはいけない。To do this, it is necessary to register the response voice once. If the system has a fixed word to recognize in advance, it is sufficient to have the data for that word in advance, but if the user registers the word to be recognized, the response voice must also be registered. .
音声認識装置では通常認識しようとする単語の辞書作成
のため、事前にその単語を複数回登録するので、同時に
応答用音声も登録すれば、応答用音声として特別に発声
することもなく、ユーザの負担も軽くなる。連続した複
数回（通常２〜４回）の発声のうち第１回目の発声は、
せき払いや″え−と”といった言葉が発声される可能性
が高く、非常に不安定である。したがって、音声区間の
切り出しがうまくいかなかったり、登録音声以外の音声
が入ってしまう可能性がある。それに比べ２回目以降の
音声は安定しているため比較的きれいな音声を登録する
ことができる。Speech recognition devices usually register the word they want to recognize multiple times in advance in order to create a dictionary, so if you also register a response voice at the same time, the user will not have to utter a special voice as a response voice. The burden will also be lighter. The first utterance of multiple consecutive utterances (usually 2 to 4 utterances) is
There is a high possibility that words such as coughing and ``um'' will be uttered, making it very unstable. Therefore, there is a possibility that the audio section may not be cut out properly or that audio other than the registered audio may be included. In comparison, the second and subsequent sounds are stable, so relatively clear sounds can be registered.
目 的
上述のように、音声認識装置においては、認識結果をア
ンサーバックする必要があるが、本発明は、このアンサ
ーバック用音声を確実に登録することのできる手段を提
供することを目的としてなされたものである。Purpose As mentioned above, in a speech recognition device, it is necessary to answer back the recognition result, and the present invention has been made for the purpose of providing a means that can reliably register this answer-back speech. It is something that
構 成
本発明は、上記目的を達成するために、入力音声に対す
る認識結果をアンサーバックする機能を有する音声認識
装置において、アンサーバックのための応答音声に認識
の辞書作成のために発声した複数回発声のうちの２回目
以降の音声を用いるようにしたことを特徴としたもので
ある。以下、本発明の実施例に基いて説明する。Configuration In order to achieve the above object, the present invention provides a speech recognition device having a function of answering back recognition results for input speech, in which multiple utterances uttered for creating a recognition dictionary are added to response speech for answerback. The feature is that the second and subsequent voices are used. Hereinafter, the present invention will be explained based on examples.
第１図は、本発明の一実施例を説明するためのブロック
図で、図中、１は音声認識装置、２は認識用辞書、３は
制御回路、４はダイヤリング装置、５は音声応答装置、
６は応答用辞書で、この第１図に示した認識システムで
は音声認識と音声応答では異なるデータを有している。FIG. 1 is a block diagram for explaining one embodiment of the present invention, in which 1 is a speech recognition device, 2 is a recognition dictionary, 3 is a control circuit, 4 is a dialing device, and 5 is a voice response Device,
6 is a response dictionary, which in the recognition system shown in FIG. 1 has different data for voice recognition and voice response.
したがって、音声入力を認識装置１と応答装置５に同時
に入力し、認識装置１の辞書作成時に制御回路３により
２回目以降のある音声が発声されたとき応答装置５に起
動をかけ応答用音声を登録する。Therefore, voice input is simultaneously input to the recognition device 1 and the response device 5, and when a certain voice is uttered from the second time onward by the control circuit 3 when the recognition device 1 creates a dictionary, the response device 5 is activated to generate a response voice. register.
第２図は、本発明の他の実施例を説明するためのブロッ
ク図で１図中、第１図に示した実施例と同様の作用をす
る部分には第１図の場合と同一の参照番号が付しである
。而して、この実施例は、認識装置ｉ２１と応答袋＠５
において同一データを用いるもので、入力音声は認識装
置１に入力し複数回の発声により認識用辞書２を作成し
、このとき２回目以降の発声のうち１つを応答用辞書６
として登録する。FIG. 2 is a block diagram for explaining another embodiment of the present invention. In FIG. 1, the same reference numerals as in FIG. It is numbered. Therefore, in this embodiment, the recognition device i21 and the response bag @5
The input speech is input to the recognition device 1 and a recognition dictionary 2 is created by uttering it multiple times.At this time, one of the second and subsequent utterances is used as the response dictionary 6.
Register as.
効 果
以上の説明から明らかなように１本発明によると音声応
答用音声を登録する場合、音声認識の登録用音声の２回
目以降の音声を用いるようにしたので、きれいな応答用
音声を登録することができる。Effects As is clear from the above explanation, according to the present invention, when registering a response voice, the second and subsequent voices of the voice recognition registration voice are used, so that a clear response voice can be registered. be able to.
第１図及び第２図は、それぞれ本発明の詳細な説明する
ためのブロック図である。
１・・・音声認識装置、２・・・認識用辞書、３・・・
制御回路、４・・・ダイヤリング装置、５・・・音声応
答装置、６・・・応答用辞書。
第１図
第２図1 and 2 are block diagrams for explaining the present invention in detail, respectively. 1... Speech recognition device, 2... Dictionary for recognition, 3...
Control circuit, 4...Dialing device, 5...Voice response device, 6...Response dictionary. Figure 1 Figure 2
Claims (1)
能を有する音声認識装置において、アンサーバックのた
めの応答音声に認識の辞書作成のために発声した複数回
発声のうちの２回目以降の音声を用いるようにしたこと
を特徴とする音声認識装置。1. In a speech recognition device that has a function of answering back the recognition results for input speech, it is possible to use the second and subsequent utterances of the multiple utterances uttered to create a recognition dictionary as the response voice for answerback. A voice recognition device characterized by:
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP63022446A JPH01197795A (en) | 1988-02-02 | 1988-02-02 | Voice recognizing device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP63022446A JPH01197795A (en) | 1988-02-02 | 1988-02-02 | Voice recognizing device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| JPH01197795A true JPH01197795A (en) | 1989-08-09 |
Family
ID=12082938
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| JP63022446A Pending JPH01197795A (en) | 1988-02-02 | 1988-02-02 | Voice recognizing device |
Country Status (1)
| Country | Link |
|---|---|
| JP (1) | JPH01197795A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2022159294A (en) * | 2019-08-08 | 2022-10-17 | アイリスオーヤマ株式会社 | Lighting device |
-
1988
- 1988-02-02 JP JP63022446A patent/JPH01197795A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2022159294A (en) * | 2019-08-08 | 2022-10-17 | アイリスオーヤマ株式会社 | Lighting device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20010056345A1 (en) | Method and system for speech recognition of the alphabet | |
| US6397182B1 (en) | Method and system for generating a speech recognition dictionary based on greeting recordings in a voice messaging system | |
| JPS60247697A (en) | Voice recognition responder | |
| JPS597998A (en) | Continuous voice recognition equipment | |
| JP2003029776A (en) | Voice recognition device | |
| JPH01197795A (en) | Voice recognizing device | |
| JPS63149699A (en) | Voice input/output device | |
| JP2002101203A (en) | Speech processing system, speech processing method and storage medium storing the method | |
| JPH11175093A (en) | Method for recognizing/confirming/responding voice | |
| JP2656234B2 (en) | Conversation voice understanding method | |
| JPH0122635B2 (en) | ||
| JPH01179198A (en) | Rejecting system | |
| JPS61239358A (en) | Documentation system by voice input | |
| JPH02103599A (en) | Voice recognizing device | |
| JPS62131298A (en) | Voice input/output unit | |
| JP2002196789A (en) | Speech interactive device | |
| JPH04116699A (en) | Voice synthesizer | |
| JPS63260253A (en) | Audio response system | |
| JPS59107391A (en) | Utterance training apparatus | |
| JPS59195299A (en) | Sepecific speaker's voice recognition equipment | |
| JPS62299997A (en) | Interactive type voice input/output unit | |
| JPH04301695A (en) | Dictionary control system for speech recognition device | |
| JP2014021425A (en) | Speech recognition system and integrated circuit device | |
| JP2005148764A (en) | Method and device for speech recognition interaction | |
| JPS63229496A (en) | Pattern updating system for voice recognition |