JPS6227719B2 - - Google Patents
Info
- Publication number
- JPS6227719B2 JPS6227719B2 JP55046232A JP4623280A JPS6227719B2 JP S6227719 B2 JPS6227719 B2 JP S6227719B2 JP 55046232 A JP55046232 A JP 55046232A JP 4623280 A JP4623280 A JP 4623280A JP S6227719 B2 JPS6227719 B2 JP S6227719B2
- Authority
- JP
- Japan
- Prior art keywords
- voice
- voice response
- subscriber
- switch
- voice recognition
- 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
Links
- 230000004044 response Effects 0.000 claims description 44
- 238000000034 method Methods 0.000 claims description 14
- 238000010586 diagram Methods 0.000 description 6
- 230000006870 function Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 206010042008 Stereotypy Diseases 0.000 description 1
- 230000000593 degrading effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000011017 operating method Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 230000001755 vocal effect Effects 0.000 description 1
Description
【発明の詳細な説明】
本発明は電話交換網を介して行なわれる銀行電
話サービス（振込照合等）などに使用される音声
応答方式に関するものである。DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a voice response system used for bank telephone services (such as bank transfer verification) performed via a telephone exchange network.
従来の音声応答方式のブロツク図を第１図に示
す。図において１は音声応答装置、２は回線装
置、３は音声応答トランク、４はプツシユホン信
号受信器、５は磁気バブルや半導体メモリ等から
構成された音声フアイルメモリ、６は音声フアイ
ルメモリ５に記憶されている語彙を取り出し編集
する音声スイツチ部、７は制御装置、８は電話機
端末である加入者、９は電話交換機を示す。 A block diagram of a conventional voice response system is shown in FIG. In the figure, 1 is a voice response device, 2 is a line device, 3 is a voice response trunk, 4 is a pushphone signal receiver, 5 is a voice file memory composed of magnetic bubbles, semiconductor memory, etc., and 6 is a memory stored in the voice file memory 5. 7 is a control device, 8 is a subscriber who is a telephone terminal, and 9 is a telephone exchange.
加入者８が例えば銀行電話サービスを要求する
場合、そのサービス番号をダイヤル（この場合プ
ツシユホン）すると周知のように電話交換機９を
通して、該電話交換機９に接続されている音声応
答装置１の回線装置２を捕捉する。回線装置２は
図で示すように音声応答トランク３とPB受信器
４が並列の形で構成されており、この回線装置２
がサービス量に応じた数だけ設けてある。該回線
装置が捕捉されると制御装置７が起動され、その
制御で今捕捉された回線装置２は音声スイツチ部
６を通して音声フアイルメモリ５に接続される。
この音声スイツチ部６は周知のように時分割スイ
ツチでできており、制御装置７によるプログラム
制御で音声フアイルメモリ５に記憶されている語
彙をとり出し必要な文章に編集すべくそのスイツ
チを開閉する。その動作手順の例を第２図、第３
図に示す。 When a subscriber 8 requests a bank telephone service, for example, he dials the service number (in this case, a pushphone number) and, as is well known, calls the line device 2 of the voice response device 1 connected to the telephone exchange 9 through a telephone exchange 9. to capture. As shown in the figure, the line device 2 is configured with a voice response trunk 3 and a PB receiver 4 in parallel.
There are a number of locations depending on the amount of service. When the line device is captured, the control device 7 is activated, and under its control, the line device 2 that has just been captured is connected to the audio file memory 5 through the audio switch unit 6.
As is well known, this voice switch section 6 is made up of a time division switch, and under program control by a control device 7, the switch is opened and closed in order to take out the vocabulary stored in the voice file memory 5 and edit it into the necessary sentences. . Examples of the operating procedure are shown in Figures 2 and 3.
As shown in the figure.
第２図において、S1，S2…S10は音声スイツチ
部６のスイツチのスイツチゲートであるクロスポ
イントを示し、該クロスポイントは回線装置２と
音声フアイルメモリ５に記憶されている各語彙に
対応している。即ち図の例では回線装置２に対応
し、音声フアイルメモリ５の「アナタノ」という
語彙に対応したクロスポイントをS1とし「ヨキン
ザンダカハ」という語彙に対応したクロスポイン
トをS2、以下同様にS3…S10と示してある。従つ
て例えば第３図に示すように、「あなたの預金残
高は××円です。有難うございました」という文
章を編集するには、制御装置７からのプログラム
制御で時分割スイツチである音声スイツチ部６の
スイツチのクロスポイントS1，S2…S10のうち
S1，S2，S3，S6，S7，S9を順次閉じて行けばよ
い。第３図のもう一つの文章の例も同様である。
このようにして音声スイツチ部６で編集された文
章は音声応答トランク３を通して音声出力として
加入者８にアナウンスされる。しかしながら一般
にこのようなサービスは音声応答だけでなく加入
者８からの情報も必要である。例えば銀行電話サ
ービスであればまずは「加入者番号と暗証番号を
どうぞ」とアナウンスされ、加入者８はその番号
情報を送らなければならない。その番号情報は現
在はプツシユホン（PB電話機）を使うことによ
つて行なわれている。即ち音声応答装置１からの
「番号をどうぞ」というアナウンスに対する答の
番号情報はPB電話機８の釦を押すことにより音
声応答装置１に伝えられる。つまりそのPB信号
を受信する受信器が必要であり、それが回線装置
２に設けられているPB受信器４である。該PB受
信器４で受信された番号情報は制御装置７で照合
され、次の手順へと進む。暗証番号や加入者コー
ド等すべてPB信号で同様に行なわれ、これらは
音声応答装置１の制御装置７に予め組込まれたプ
ログラムの手順に従つてアナウンスと、加入者か
らの情報とが会話的になされる。 In FIG. 2, S 1 , S 2 . . . S 10 indicate cross points that are switch gates of the audio switch unit 6. Compatible. That is, in the example shown in the figure, the cross point corresponding to the line device 2 and the word "anatano" in the voice file memory 5 is S1 , the cross point corresponding to the word "Yokin Zandakaha" is S2 , and the same goes for S3. ...It is marked as S 10 . Therefore, for example, as shown in Figure 3, in order to edit the sentence "Your bank balance is xx yen. Thank you very much," the voice switch part, which is a time division switch, is programmed from the control device 7. 6 switch cross points S 1 , S 2 ... out of S 10
S 1 , S 2 , S 3 , S 6 , S 7 , and S 9 may be closed in sequence. The same is true for the other sentence example in Figure 3.
The text edited by the voice switch section 6 in this manner is announced to the subscriber 8 as a voice output through the voice response trunk 3. However, such services generally require information from the subscriber 8 as well as a voice response. For example, in the case of a bank telephone service, an announcement is first made saying, ``Please enter your subscriber number and PIN number,'' and the subscriber 8 must send the number information. The number information is currently obtained by using a pushphone (PB telephone). That is, the number information in response to the announcement "Please give me your number" from the voice response device 1 is transmitted to the voice response device 1 by pressing a button on the PB telephone 8. In other words, a receiver is required to receive the PB signal, and that is the PB receiver 4 provided in the line device 2. The number information received by the PB receiver 4 is verified by the control device 7, and the process proceeds to the next step. PIN numbers, subscriber codes, etc. are all processed in the same way using PB signals, and announcements and information from the subscriber are communicated conversationally according to the procedure of a program pre-installed in the control device 7 of the voice response device 1. It will be done.
以上の説明のようにこのような音声応答サービ
スにあつては加入者側からの情報は音声帯域内信
号をもつPB電話機でしかできない。それは加入
者８が音声応答装置１に接続された後は音声帯域
内信号（300Hz〜3400Hz）しか利用できないから
である。即ち、従来このようなサービスはPB電
話機の加入者しか利用できなかつた。そのために
音声応答装置１の回線装置２毎にPB受信器を設
けてあつた。しかるに近年このようなサービスが
普及してくるに従い、まだ電話加入者の大部分を
占めている回転ダイヤル電話機からもこのような
サービスの利用要求が高まつてきた。またPB電
話機の加入者でも加入者からの情報入力を一々押
釦で行なうより音声を用いて直接会話を行なう方
がより便利であり、そのため、これらの要求に応
えるためにPB受信器４のほかに音声認識装置を
設備した音声応答装置の要求が高まりつつある。 As explained above, in the case of such voice response services, information from the subscriber side can only be received by PB telephones that have signals within the voice band. This is because after the subscriber 8 is connected to the voice response device 1, only signals within the voice band (300Hz to 3400Hz) are available. That is, in the past, such services were available only to subscribers of PB telephones. For this purpose, a PB receiver was provided for each line device 2 of the voice response device 1. However, as such services have become popular in recent years, there has been an increasing demand for the use of such services from rotary dial telephones, which still account for the majority of telephone subscribers. Furthermore, even for subscribers of PB telephones, it is more convenient to have a direct conversation using voice rather than inputting information by pressing a button one by one. Therefore, in order to meet these demands, in addition to the PB receiver 4, Demand for voice response devices equipped with voice recognition devices is increasing.
しかしながら音声認識装置の価格は、第１図に
示すPB受信器４より格段に高価なものである。
これを回線装置２毎に設けることは経済的に負担
が大きすぎ実現は困難である。 However, the cost of the speech recognition device is much more expensive than the PB receiver 4 shown in FIG.
Providing this for each line device 2 would be an economical burden and would be difficult to implement.
この欠点を解決するため、第１図における回線
装置２の内部に音声認識装置を組込むことをせ
ず、時分割スイツチである音声編集スイツチ部６
を拡張し、これに音声認識装置４を収容し、加入
者との会話で必要な時点に限りこの音声認識装置
４と回線装置２を時分割的に接続し、高価な音声
応答装置４を多数の回線装置２で共通使用できる
ようにすれば音声認識装置４の使用効率が向上
し、音声応答装置のサービスを落すことなく、音
声認識装置４の設備数の大巾削減が行なえる。 In order to solve this drawback, instead of incorporating a voice recognition device inside the line device 2 shown in FIG.
The voice recognition device 4 is expanded and a voice recognition device 4 is housed in it, and the voice recognition device 4 and the line device 2 are connected in a time-division manner only at the time when it is necessary for a conversation with a subscriber, and a large number of expensive voice response devices 4 are installed. If the voice recognition device 4 can be used in common with other line devices 2, the usage efficiency of the voice recognition device 4 will be improved, and the number of voice recognition devices 4 can be significantly reduced without degrading the service of the voice response device.
しかしながら、音声認識装置４の価格は通常認
識すべき言葉の数に応じて上昇するが、銀行電話
サービス用音声認識装置のように不特定利用者か
らの音声認識機能が要求されるものについては、
同一意味の言葉であつても、発声上の音の質（年
令、性別）やくせの認識が要求されるため、特に
認識語数が多い場合周知のようにこれらを認識す
るための基準音声パタンメモリが大巾に増加し、
かつ利用者の音声と基準音声メモリとの実時間処
理による照合の組合せが増加する。すなわち速い
処理装置と効率的な認識を行なうための高度な技
術を駆使したプログラムが必要になり、その結果
急激な価格上昇をきたす結果となる。 However, the price of the voice recognition device 4 usually increases depending on the number of words to be recognized, but for voice recognition devices for bank telephone services that require a voice recognition function from unspecified users,
Even if the words have the same meaning, it is necessary to recognize the quality of the vocal sounds (age, gender) and mannerisms, so when there are a large number of words to be recognized, it is well known that standard speech patterns are needed to recognize these. Memory has increased dramatically,
In addition, the number of combinations of comparisons between the user's voice and the reference voice memory through real-time processing increases. In other words, a fast processing device and a program that makes full use of advanced technology for efficient recognition are required, resulting in a rapid increase in price.
従つて前述のように回線装置と音声認識装置の
間にスイツチを設け、音声認識装置を共通使用し
て、音声認識装置の設備数軽減を計つたとしても
認識語数を少くしない限り不特定話者の音声認識
装置があまりにも高価であるため、音声応答装置
部の価格低減は十分に行なえない。 Therefore, even if a switch is installed between the line device and the speech recognition device as mentioned above and the speech recognition devices are used in common to reduce the number of speech recognition devices, unless the number of recognized words is reduced, unspecified speakers Since the voice recognition device is too expensive, it is not possible to reduce the cost of the voice response device unit sufficiently.
本発明は利用者と音声応答装置との会話手順
（第５図を使用して後述する。）に応じて少い語数
たとえば“はい、いいえ”あるいは“０〜９の数
字”程度のみ認識する安価な不特定話者用音声認
識装置を切替使用するとにより経済化を計つた音
声応答装置に関するものである。以下詳細に説明
する。 The present invention is an inexpensive method that recognizes only a small number of words, such as "yes, no" or "numbers 0 to 9," depending on the conversation procedure between the user and the voice response device (described later using FIG. 5). The present invention relates to a voice response device that is economical by switching and using voice recognition devices for unspecified speakers. This will be explained in detail below.
第４図は本発明の実施例のブロツク図であつ
て、２０は回線装置、２１は音声応答トランク、
２２はアナログ信号からデイジタル信号へ、また
デイジタル信号をアナログ信号に変更するコンバ
ータ、２３は音声デイジタル信号を一旦蓄積する
バツフアー、２４は音声スイツチ部、２５，２
６，２７，２８，３４は時分割スイツチゲート、
３５は時分割スイツチバス、２９は“はい、いい
え”程度の少い語数を認識する不特定話者用音声
認識装置、３１は制御装置、３２は音声フアイル
メモリ、３０は“０〜９”の数字だけを認識する
不特定話者用音声認識装置、３３はFS信号送出
器、８は加入者、９は電話交換機を示す。 FIG. 4 is a block diagram of an embodiment of the present invention, in which 20 is a line device, 21 is a voice response trunk,
22 is a converter for converting an analog signal into a digital signal and converting a digital signal into an analog signal; 23 is a buffer for temporarily accumulating the audio digital signal; 24 is an audio switch section; 25, 2
6, 27, 28, 34 are time division switch gates,
35 is a time division switch bus, 29 is a speech recognition device for non-specific speakers that recognizes a small number of words such as "yes, no", 31 is a control device, 32 is a voice file memory, and 30 is a number from "0 to 9". 33 is an FS signal transmitter, 8 is a subscriber, and 9 is a telephone exchange.
加入者８が音声応答サービスを要求した場合、
電話交換機９を通して回線装置２０の音声応答ト
ランク２１に接続されてくるのは従来の例の第１
図と同様である。音声応答トランク２１は捕捉さ
れると制御装置３１を起動し、該制御装置３１の
プログラム制御で音声応答トランク２１をコンバ
ータ２２、バツフアー２３を経て音声スイツチ部
２４を通し音声フアイルメモリ３２に接続する。
この音声スイツチ部２４は時分割スイツチで構成
されており、ゲートスイツチ２５は回線装置２０
に、ゲートスイツチ２６は音声フアイルメモリ３
２に、２７は音声認識装置２９に、２８は音声認
識装置３０にそれぞれ対応している。かつゲート
スイツチ２５は回線装置２０、ゲートスイツチ２
６は音声フアイルメモリに記憶されている各語彙
に対応させる。この構成は第２図で例示したクロ
スポイントS1…S10と同様である。従つて今接続
されてきた回線装置２０に対応するゲートスイツ
チ２６を閉じ、音声フアイルメモリ３２に記憶さ
れている語彙のうち今必要とする文章になる語彙
に対応するゲートスイツチ２６を順次閉じていく
ことにより応答文が編集される。一般に時分割ス
イツチを使用し、磁気デイスク等のメモリを使用
する場合、デイジタル信号で記憶し、取り出すの
が有利であることは周知であるので本実施例でも
それを前提にしている。従つてデイジタル信号で
記憶している音声フアイルメモリ３２の語彙をゲ
ートスイツチ２６の動作により音声スイツチ部２
４で編集された文章はまだデイジタル信号のまま
であるので、その文章を加入者８に音声として聞
かせるため回線装置２０にはコンバータ２２とバ
ツフアー２３を設け、前記デイジタル信号の文章
をバツフアー２３に一旦蓄積し、コンバータ２２
でデイジタル信号からアナログ信号即ち音声へ変
換して音声応答トランク２１から加入者８へ聞か
す。以上の説明は従来の例第１図の音声スイツチ
部６で音声フアイルメモリ５の語彙を編集して回
線装置２を通して音声応答することと同様であ
る。 If subscriber 8 requests voice response service,
The first one in the conventional example is connected to the voice response trunk 21 of the line equipment 20 through the telephone exchange 9.
It is similar to the figure. When the voice response trunk 21 is captured, the control device 31 is activated, and the voice response trunk 21 is connected to the voice file memory 32 through the converter 22, buffer 23, and voice switch unit 24 under program control of the control device 31.
This audio switch section 24 is composed of a time division switch, and the gate switch 25 is connected to the line device 20.
, the gate switch 26 is connected to the audio file memory 3.
2, 27 correspond to the voice recognition device 29, and 28 correspond to the voice recognition device 30, respectively. And the gate switch 25 is connected to the line device 20 and the gate switch 2.
6 corresponds to each vocabulary stored in the audio file memory. This configuration is similar to the cross points S1 ... S10 illustrated in FIG. Therefore, the gate switch 26 corresponding to the line device 20 that has just been connected is closed, and the gate switches 26 corresponding to the vocabulary that will form the sentence that is currently needed among the vocabulary stored in the audio file memory 32 are sequentially closed. The response text is edited accordingly. Generally, when a time division switch is used and a memory such as a magnetic disk is used, it is well known that it is advantageous to store and retrieve digital signals, and this embodiment is also based on this. Therefore, the vocabulary stored in the audio file memory 32 as a digital signal is transferred to the audio switch unit 2 by the operation of the gate switch 26.
Since the text edited in step 4 is still a digital signal, the line device 20 is provided with a converter 22 and a buffer 23 in order to make the text audible to the subscriber 8, and the text in the digital signal is sent to the buffer 23. Once stored, the converter 22
The digital signal is converted into an analog signal, ie, voice, and is heard by the subscriber 8 through the voice response trunk 21. The above explanation is similar to the conventional example of editing the vocabulary in the voice file memory 5 using the voice switch section 6 shown in FIG. 1 and giving voice responses through the line device 2.
本実施例は音声スイツチ部２４に認識語数が多
い反面高価な不特定話者用認識装置を収容するか
わりに、利用者と音声応答装置との会話手順に応
じて、少い言葉たとえば“はい、いいえ”あるい
は“０〜９”の数字程度しか認識できない安価な
不特定話者用音声認識装置２９，３０を切替使用
することにより音声応答装置の音声認識部の経済
化を計るものである。 In this embodiment, instead of accommodating in the voice switch section 24 a recognition device for non-specified speakers which can recognize a large number of words but is expensive, the voice switch section 24 can recognize a small number of words, such as "Yes," according to the conversation procedure between the user and the voice response device. The purpose is to make the voice recognition section of the voice response device economical by switching between inexpensive voice recognition devices 29 and 30 for unspecified speakers that can only recognize numbers such as "No" or "0 to 9."
即ち例えば銀行電話サービスの場合、加入者８
から接続されてくるとまずは「こちらはABC銀
行です加入者番号と暗証番号をどうぞ」とアナウ
ンスするのが通例である。従つて音声応答装置と
しては前述したような動作でそのアナウンスを行
なうがこのとき（アナウンス送出時でもアナウン
ス直後でもよい）制御装置３１のプログラム制御
で、音声スイツチ部２４のゲートスイツチ２８を
閉じ音声認識装置３０をも今接続されてきた回線
装置２０にゲートスイツチ２５により接続状態に
する。そうすれば前述のアナウンス後加入者８が
音声で加入者番号を「２・３・４・５」のように
数字音で発声すればそれを音声認識装置３０で受
信する。次に音声応答装置から「加入者番号は
２・３・４・５ですね。正しければはいと云つて
下さい。誤つていたらいいえと云つて下さい」と
アナウンスし、「はいまたはいいえ」を受信す
る。「はい」を受信すると次に「暗証番号をどう
ぞ」とアナウンスする。加入者が「はい」と発声
する。この「はい，いいえ」の音声は音声認識装
置２９で受信し、制御装置３１で加入者番号と照
合番号を照合する。これらの手順は第５図の通り
である。第５図中VRDと表示してあるのは音声
認識装置の略号である。このような手順に従つて
音声認識装置２９，３０を必要な時のみ切替接続
して使用することは周知のようにプログラム制御
の装置にあつてはプログラムを予めその手順に従
つて設定しておくことにより容易に実現できる。 For example, in the case of bank telephone service, subscriber 8
When someone connects you, it is customary to first announce, ``This is ABC Bank. Please enter your subscriber number and PIN.'' Therefore, the voice response device makes the announcement by the operation described above, but at this time (either at the time of sending the announcement or immediately after the announcement) the gate switch 28 of the voice switch section 24 is closed under the program control of the control device 31 and the voice recognition is performed. The device 30 is also connected to the currently connected line device 20 by the gate switch 25. Then, after the above-mentioned announcement, if the subscriber 8 vocally utters the subscriber number in numeric sounds such as "2, 3, 4, 5", the voice recognition device 30 will receive it. Next, the voice response device announces, ``The subscriber number is 2, 3, 4, 5. If it is correct, please say yes. If it is incorrect, please say no.'' and receives ``Yes or No.'' do. If "Yes" is received, the next step is to announce "Please enter your PIN number." The subscriber says "Yes". This "yes, no" voice is received by the voice recognition device 29, and the control device 31 verifies the subscriber number and verification number. These steps are shown in FIG. In FIG. 5, VRD is an abbreviation for voice recognition device. It is well known that the voice recognition devices 29 and 30 are switched and connected only when necessary according to such a procedure, and in the case of a program-controlled device, the program is set in advance according to the procedure. This can be easily achieved by
以上の説明から明らかなように、本発明では認
識語数が多い反面極めて高価な不特定話者用音声
認識装置を使用することなく第５図のように利用
者（加入者）と音声応答装置との会話手順に応じ
て、少い言葉たとえば「はい、いいえ」あるいは
「０，１〜９」の数字程度のみ認識する安価な不
特定話者用音声認識装置を切替使用することによ
り、音声応答装置の最も高価な部分である音声認
識部の経済化を計るものである。 As is clear from the above explanation, in the present invention, the number of words to be recognized is large, but the voice recognition device is able to communicate with the user (subscriber) and the voice response device as shown in FIG. 5 without using an extremely expensive speech recognition device for non-specific speakers. By switching and using an inexpensive speech recognition device for non-specific speakers that recognizes only a few words, such as “yes, no” or numbers such as “0, 1 to 9,” depending on the conversation procedure, the voice response device The aim is to make the voice recognition section, the most expensive part of the system, more economical.
以上の説明は音声認識装置２９，３０の収容位
置として時分割スイツチ部２４について述べた
が、２４は第５図に示した各手順において、音声
認識装置２９，３０が高速で切替ればよいのであ
つて、たとえば電子素子による空間分割形スイツ
チを使用しても当然のことながら同等の効果が得
られる。 In the above explanation, the time division switch section 24 has been described as a housing location for the voice recognition devices 29 and 30, but the time division switch section 24 can be used as long as the voice recognition devices 29 and 30 are switched at high speed in each procedure shown in FIG. Naturally, the same effect can be obtained even if a space division type switch using an electronic element is used, for example.
また第６図のように回線装置２の中に含まれる
音声トランク３に音声認識装置２９，３０を並列
接続しても、音声認識装置２９，３０は認識語数
が少い安価なものであるから、第４図のように多
数の回線装置との共用化を計らなくても音声認識
部の経済化は可能である。 Furthermore, even if the voice recognition devices 29 and 30 are connected in parallel to the voice trunk 3 included in the line device 2 as shown in FIG. 6, the voice recognition devices 29 and 30 are inexpensive devices that can recognize only a small number of words. , it is possible to make the voice recognition section economical without having to share it with many line devices as shown in FIG.
以上説明したように音声認識装置を持つた音声
応答装置において、限られた少数語より認識する
機能しかもたない安価な不特定話者用音声認識装
置を、第５図のような各手順に応じて切替使用す
ることにより、音声応答システムにおける不特定
話者用音声認識部の価格を大巾に低減できる。 As explained above, in a voice response device equipped with a voice recognition device, an inexpensive voice recognition device for non-specific speakers, which only has the function of recognizing a limited number of words, can be used according to each procedure as shown in Figure 5. By switching and using the system, the cost of the speaker-independent voice recognition unit in the voice response system can be significantly reduced.
このような音声応答サービスは銀行電話サービ
スは勿論、種々の予約業務、商品発注業務等に使
用されるが従来これらはプツシユホン加入者のみ
に限られていた。本発明は前述したように音声で
の加入者からの情報をも経済的に受信できる音声
応答方式を実現したため、一般の回転ダイヤル電
話加入者、データテレホン使用者等からもこのサ
ービスが利用できるようになつた。従つて加入者
即ち利用者の拡大が図れるとともに、今後各分野
で利用される音声応答サービスに対し経済的な方
式が提供できる。 Such voice response services are used not only for bank telephone services but also for various reservation operations, product ordering operations, etc., but in the past, these services were limited only to pushphone subscribers. As mentioned above, the present invention has realized a voice response system that can economically receive information from subscribers by voice, so that general rotary dial telephone subscribers, data telephone users, etc. can also use this service. It became. Therefore, it is possible to increase the number of subscribers, that is, users, and to provide an economical method for voice response services that will be used in various fields in the future.
第１図は従来の音声応答方式のブロツク図、第
２図は音声スイツチ部の原理図、第３図は文章編
集例、第４図は本発明の実施例のブロツク図、第
５図は実施例の動作手順例、第６図は本発明の他
の実施例のブロツク図である。
８……加入者、９……電話交換機、２０……回
線装置、２１……音声応答トランク、２２……コ
ンバータ、２３……バツフアー、２４……音声ス
イツチ部、２５，２６２７，２８，３４……ゲー
トスイツチ、２９，３０……音声認識装置、３１
……制御装置、３２……音声フアイルメモリ、３
３……FS信号送出器、３５……スイツチバス。
Figure 1 is a block diagram of a conventional voice response system, Figure 2 is a principle diagram of the voice switch section, Figure 3 is an example of text editing, Figure 4 is a block diagram of an embodiment of the present invention, and Figure 5 is an implementation example. EXAMPLE OF OPERATION PROCEDURE FIG. 6 is a block diagram of another embodiment of the present invention. 8...Subscriber, 9...Telephone exchange, 20...Line equipment, 21...Voice response trunk, 22...Converter, 23...Buffer, 24...Voice switch section, 25, 2627, 28, 34... ...gate switch, 29,30...speech recognition device, 31
...control device, 32 ...audio file memory, 3
3...FS signal transmitter, 35...Switch bus.
Claims (1)
て、該方式の利用者との会話のうち音声認識すべ
き言葉毎に該言葉のみを認識する音声認識装置を
複数個設け、前記会話の手順に従つてその時必要
な該音声認識装置を切替え接続することを特徴と
する音声応答方法。1. In a voice response system equipped with a voice recognition device, a plurality of voice recognition devices are installed to recognize only the words to be voice recognized in a conversation with a user of the system, and each word is recognized at that time according to the procedure of the conversation. A voice response method characterized by switching and connecting necessary voice recognition devices.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP4623280A JPS56143495A (en) | 1980-04-10 | 1980-04-10 | Voice responding system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP4623280A JPS56143495A (en) | 1980-04-10 | 1980-04-10 | Voice responding system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| JPS56143495A JPS56143495A (en) | 1981-11-09 |
| JPS6227719B2 true JPS6227719B2 (en) | 1987-06-16 |
Family
ID=12741362
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| JP4623280A Granted JPS56143495A (en) | 1980-04-10 | 1980-04-10 | Voice responding system |
Country Status (1)
| Country | Link |
|---|---|
| JP (1) | JPS56143495A (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS59193500A (en) * | 1983-04-18 | 1984-11-02 | 富士通株式会社 | Non-specification-specification changeover system for voice recognition unit |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS5521035A (en) * | 1978-07-31 | 1980-02-14 | Tokyo Shibaura Electric Co | Control system using voice |
-
1980
- 1980-04-10 JP JP4623280A patent/JPS56143495A/en active Granted
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS5521035A (en) * | 1978-07-31 | 1980-02-14 | Tokyo Shibaura Electric Co | Control system using voice |
Also Published As
| Publication number | Publication date |
|---|---|
| JPS56143495A (en) | 1981-11-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US5809464A (en) | Apparatus for recording speech for subsequent text generation | |
| US6687673B2 (en) | Speech recognition system | |
| JP2957862B2 (en) | Communication system and communication method | |
| US5594784A (en) | Apparatus and method for transparent telephony utilizing speech-based signaling for initiating and handling calls | |
| JP3168033B2 (en) | Voice telephone dialing | |
| CA2172748C (en) | Digital secretary | |
| US6493670B1 (en) | Method and apparatus for transmitting DTMF signals employing local speech recognition | |
| US6601031B1 (en) | Speech recognition front end controller to voice mail systems | |
| EP0788268A1 (en) | Process for interaction between a telephone and its user | |
| JP3204632B2 (en) | Voice dial server | |
| US6256611B1 (en) | Controlling a telecommunication service and a terminal | |
| KR101367722B1 (en) | Method for communicating voice in wireless terminal | |
| JPH08163252A (en) | Pbx/computer interlock system | |
| US6845356B1 (en) | Processing dual tone multi-frequency signals for use with a natural language understanding system | |
| JPS6227719B2 (en) | ||
| JPS61144157A (en) | Sound dial device | |
| US20040037399A1 (en) | System and method for transferring phone numbers during a voice call | |
| JPH07175495A (en) | Voice recognition system | |
| JP3278595B2 (en) | mobile phone | |
| KR940011237B1 (en) | Phonetic dialing method and apparatus | |
| JPS5860863A (en) | Transmission system for tone for indicating premission to voice | |
| JPH10190842A (en) | Speech interactive system | |
| JP2000151827A (en) | Telephone voice recognizing system | |
| JPH06303320A (en) | Pbx device | |
| JP2000122692A (en) | Speech recognizer |