JPH07177217A - Dial tone detecting method for automatic answering telephone - Google Patents

Dial tone detecting method for automatic answering telephone

Info

Publication number
JPH07177217A
JPH07177217A JP5318482A JP31848293A JPH07177217A JP H07177217 A JPH07177217 A JP H07177217A JP 5318482 A JP5318482 A JP 5318482A JP 31848293 A JP31848293 A JP 31848293A JP H07177217 A JPH07177217 A JP H07177217A
Authority
JP
Japan
Prior art keywords
block
blocks
zero
dial tone
voiced
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
Application number
JP5318482A
Other languages
Japanese (ja)
Inventor
Koichiro Kubo
宏一郎 久保
Yoshio Ogawa
良夫 小川
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.)
Nippon Telegraph and Telephone Corp
Original Assignee
Nippon Telegraph and Telephone Corp
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 Nippon Telegraph and Telephone Corp filed Critical Nippon Telegraph and Telephone Corp
Priority to JP5318482A priority Critical patent/JPH07177217A/en
Publication of JPH07177217A publication Critical patent/JPH07177217A/en
Pending legal-status Critical Current

Links

Abstract

PURPOSE:To easily and surely detect a dial tone following an answer message without depending upon the kind of an equipment by dividing an audible signal into time-series blocks and inspecting the voiced/voiceless states of the blocks and zero-crossing frequency variation. CONSTITUTION:A speech circuit 5 divides the audible tone from an opposite- party automatic answering telephone 1 into the time-series blocks. A voiced/ voiceless decision part 101 inspects mean electric power, block by block, through a mean electric power detection part 106 and stores decision data on a voiced state in an FIFO buffer 10 when, for example, a specific threshold value is exceeded. A zerocrossing frequency count part 103 counts the zero-crossing frequency of the signal tone, block by block, and stores the difference from a last block in an FIFO buffer 104. A discrimination condition decision part 105 analyzes the data stored in the FIFO buffers 102 and 104 and decides that the dial tone is received from the opposite-party automatic answering telephone 1 when a specific number of blocks which are in the voiced state and less than in zero-crossing frequency than specified succeed. A voice information sending-out part 6 sends out specific voice information.

Description

【発明の詳細な説明】Detailed Description of the Invention

【0001】[0001]

【産業上の利用分野】この発明は、留守番電話の発信音
検出方法に関する。
BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a dial tone detection method for an answering machine.

【0002】[0002]

【従来の技術】留守番電話の発信音検出方法の従来例を
説明する。音声メール或は各種の通知、その他の音声情
報を電話回線を介して電話加入者に対して自動的に伝送
する際に、加入者側の端末装置が留守番電話である場
合、留守番電話に予め登録されている応答メッセージの
再生が終了し、次いで発信音の出力がなされた後に音声
情報を当方から送出する必要がある。
2. Description of the Related Art A conventional example of a dial tone detection method for an answering machine will be described. When automatically transmitting voice mail or various notifications and other voice information to a telephone subscriber through a telephone line, if the terminal device on the subscriber side is an answering machine, pre-register on the answering machine. It is necessary to send the voice information from me after the reproduction of the response message being performed is finished and then the dial tone is output.

【0003】音声情報を留守番電話に対して送出する
際、従来は、回線接続後直ちに音声情報を送出する方法
(特開平4−26262号参照)、或は留守番電話から
発信される応答メッセージの音声区間が終了したことを
検出し、これに基づいて音声情報を送出する方法(特開
平2−184163号参照)を採用していた。しかし、
回線接続後直ちに音声情報を送出する方法は、留守番電
話の応答メッセージの再生期間と送達すべき音声情報の
再生伝送期間とが重なるところから、相手側に当方の音
声情報を正しく伝送することができない場合がある。
When sending voice information to an answering machine, conventionally, a method of sending voice information immediately after connecting a line (see Japanese Patent Laid-Open No. 26262/1992) or a voice of a response message sent from an answering machine is used. The method of detecting the end of the section and transmitting the audio information based on this is adopted (see Japanese Patent Laid-Open No. 2-184163). But,
The method of transmitting voice information immediately after connecting the line cannot correctly transmit the voice information of the other party because the reproduction period of the answering message of the answering machine and the reproduction transmission period of the voice information to be delivered overlap. There are cases.

【0004】そして、応答メッセージの音声区間が終了
したことを検出してこれに依存する方法においては、相
手側の応答メッセージの音量、相手側における周囲の雑
音その他の条件が様々であるので、これらの相違を吸収
して音声区間の終了を正確に検出する処理をしなければ
ならないが、この処理は大変に複雑なものとなる。応答
メッセージの音声区間の終了を検出することができたも
のとしても、応答メッセージに引続いて相手側から発信
音が送出される以前に当方から音声情報が送出された場
合は、留守番電話の機種によっては実際には留守番電話
の録音が開始されないにも関わらず当方から音声情報が
送出されたことになり、結局、音声情報の先頭部分が録
音から欠落するという問題が生ずる。
In the method of detecting and depending on the end of the voice section of the response message, the volume of the response message of the other party, ambient noise on the other party, and other conditions are various. It is necessary to absorb the difference between the two and accurately detect the end of the voice section, but this processing becomes very complicated. Even if it is possible to detect the end of the voice section of the response message, if the voice information is sent from me before the dial tone is sent from the other party following the response message, the answering machine model In some cases, the voice information is transmitted from me even though the recording of the answering machine is not actually started, which eventually causes a problem that the head portion of the voice information is missing from the recording.

【0005】以上の通り、従来の方法においては、留守
番電話が録音を開始するタイミングを正確に検出するこ
とができないために、音声情報を相手側に欠落なく伝送
することができない場合が生じた。
As described above, in the conventional method, there is a case in which the voice information cannot be transmitted to the other party without omission because the timing when the answering machine starts recording cannot be accurately detected.

【0006】[0006]

【発明が解決しようとする課題】この発明は、留守番電
話の発信音を正確且つ、簡易に検出し、これにより、留
守番電話の応答メッセージの長短、留守番電話の機種に
依存せず、的確なタイミングにより留守番電話に対して
音声情報を伝送することができる留守番電話の発信音検
出方法を提供するものである。
DISCLOSURE OF THE INVENTION The present invention detects the dial tone of an answering machine accurately and easily, and by this means, it does not depend on the length of the answering message of the answering machine or the model of the answering machine, and the proper timing. Provides a method for detecting the dial tone of an answering machine, which is capable of transmitting voice information to the answering machine.

【0007】[0007]

【課題を解決するための手段】留守番電話から応答メッ
セージに引き続いて発信される発信音を検出する留守番
電話の発信音検出方法において、留守番電話から送出さ
れる可聴音を複数の時系列ブロックに分割し、各時系列
ブロックが有音であるか或は無音であるかを有音/無音
時系列データとして出力し、各時系列ブロックの零交差
数の変動を零交差数変動時系列データとして出力し、有
音/無音時系列データおよび零交差数変動時系列データ
のブロック毎の変動に基づいて、可聴音の時系列ブロッ
クが一定ブロック数に亘り連続して有音であると共に、
時系列ブロックの零交差数の変動が或る一定値よりも小
である場合、その連続するブロックにおける可聴音は留
守番電話から発信される発信音であると判定する留守番
電話の発信音検出方法を構成した。
[Means for Solving the Problems] In a dial tone detection method for an answering machine, which detects a dial tone transmitted from an answering machine following a response message, an audible tone transmitted from the answering machine is divided into a plurality of time series blocks. Then, whether each time-series block is voiced or silent is output as voiced / silent time-series data, and the fluctuation of the zero-crossing number of each time-series block is output as zero-crossing number fluctuation time-series data. However, based on the fluctuation of each block of the voiced / silent time-series data and the zero-crossing number fluctuation time-series data, the audible time-series blocks are continuously voiced over a certain number of blocks, and
If the variation of the number of zero crossings in a time series block is smaller than a certain fixed value, the audible sound in the consecutive blocks is determined to be the dial tone transmitted from the answering machine. Configured.

【0008】[0008]

【実施例】この発明の実施例を図1を参照して説明す
る。図1において、1は音声情報を電話回線を介して伝
送されるべき留守番電話、2は公衆電話回線網であり、
留守番電話1は公衆電話回線網2に接続している。3は
留守番電話1の応答を検出する応答検出部、4は発呼部
である。
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT An embodiment of the present invention will be described with reference to FIG. In FIG. 1, 1 is an answering machine for transmitting voice information via a telephone line, 2 is a public telephone line network,
The answering machine 1 is connected to the public telephone line network 2. Reference numeral 3 is a response detecting section for detecting a response from the answering machine 1, and 4 is a calling section.

【0009】5は通話回路、6は音声情報送出部、7は
制御部、100はこの発明を実施するに際して使用され
る発信音検出装置である。発信音検出装置100は有音
/無音判定部101、有音/無音判定部101の判定結
果を過去の一定時間に亘って格納するFIFOバッファ
102、零交差数係数部103、零交差数計数部103
の過去の一定時間に亘る変動量である差分を格納するF
IFOバッファ104、識別条件判定部105、および
平均電力検出部106より成る。
Reference numeral 5 is a communication circuit, 6 is a voice information sending unit, 7 is a control unit, and 100 is a dial tone detection device used in implementing the present invention. The dial tone detection device 100 includes a voice / silence determination unit 101, a FIFO buffer 102 that stores the determination results of the voice / silence determination unit 101 for a certain past time, a zero-crossing number coefficient unit 103, and a zero-crossing number counting unit. 103
F, which stores the difference that is the amount of change over the past certain time
It is composed of an IFO buffer 104, an identification condition determination unit 105, and an average power detection unit 106.

【0010】ここで、公衆電話回線網2に接続される留
守番電話1に対して、発呼部4を介して自動的に発呼動
作を行ない、応答検出部3によりその応答を検出する。
応答を検出した後、当該留守番電話1から送出される応
答メッセージを含む可聴音を受信解析し、可聴音に引続
いて送出される発信音を検出した後、当該留守番電話1
に対して予め設定した音声情報を当方から送出するもの
である。
Here, the answering machine 1 connected to the public telephone line network 2 automatically performs a calling operation via the calling section 4, and the response detecting section 3 detects the response.
After the response is detected, the audible sound including the response message transmitted from the answering machine 1 is received and analyzed, and the dial tone transmitted subsequently to the audible sound is detected.
The voice information set in advance is transmitted from me.

【0011】以下、この発明を実施するに際して使用さ
れる発信音検出装置100について説明する。この発信
音検出装置100は、留守番電話1の応答を検出した
後、留守番電話1から出力される可聴音を通話回路5に
より複数の時系列ブロックに分割し、これら時系列ブロ
ックを平均電力検出部106および零交差数計数部10
3のそれぞれに入力する。有音/無音判定部101にお
いては、平均電力検出部106の検出結果であるブロッ
ク内平均電力と零交差数計数部103の計数結果である
ブロック内零交差数とに基づいて当該ブロックの有音/
無音を判定する。例えば、ブロック内平均電力について
高低2つの閾値を設け、ブロック内平均電力が高閾値よ
りも大きな値であれば当該ブロックは有音、低閾値より
も小さければ無音、高低2つの閾値の間であればブロッ
ク内零交差数の大小により当該ブロックの有音/無音を
決定する。その判定結果を履歴として過去の一定ブロッ
クに亘りFIFOバッファ102に逐次格納する。この
判定結果を図示すると、図2に示される通りになる。
The dial tone detection device 100 used for implementing the present invention will be described below. The dial tone detection device 100 detects the response of the answering machine 1, and then divides the audible sound output from the answering machine 1 into a plurality of time series blocks by the communication circuit 5, and the time series blocks are divided into average power detection units. 106 and zero-crossing number counting unit 10
Fill in each of the three. In the sound / silence determining unit 101, the sound of the block is detected based on the in-block average power which is the detection result of the average power detecting unit 106 and the in-block zero crossing number which is the counting result of the zero crossing number counting unit 103. /
Determine silence. For example, two thresholds of high and low are set for the average power in the block, and if the average power in the block is a value higher than the high threshold, the block is voiced, if the average power in the block is lower than the low threshold, it is silent, and between the high and low thresholds. For example, sound / silence of the block is determined by the number of zero crossings in the block. The determination result is sequentially stored as a history in the FIFO buffer 102 over a certain past block. The result of this determination is illustrated in FIG.

【0012】零交差数計数部103においては、入力さ
れた当該ブロック内の零交差数を計数し、前ブロックの
零交差数との間の差分を計数してFIFOバッファ10
4に供給し、この差分をFIFOバッファ104に履歴
として過去の一定ブロックに亘り逐次格納する。この蓄
積される零交差数波形は図2に示される通りになる。識
別条件判定部105においては、FIFOバッファ10
2に蓄積される有音/無音の判定履歴とFIFOバッフ
ァ104に蓄積される零交差数の差分履歴を分析する。
FIFOバッファ102に蓄積される有音/無音の判定
履歴が一定ブロック数に亘って連続して有音であると共
に、図2に示される発信音部分の如くFIFOバッファ
104に蓄積される零交差数の差分履歴が当該ブロック
において連続して閾値より小さい場合、入力された可聴
音は留守番電話1の発信音である旨の応答を判定結果と
して制御部7に出力する。
In the zero-crossing number counting unit 103, the number of zero-crossings in the input block is counted, and the difference from the number of zero-crossings in the preceding block is counted to determine the FIFO buffer 10.
4, and the difference is sequentially stored in the FIFO buffer 104 as a history over a certain past block. The accumulated zero-crossing number waveform is as shown in FIG. In the identification condition determination unit 105, the FIFO buffer 10
The voice history / non-voice determination history accumulated in 2 and the difference history of the number of zero crossings accumulated in the FIFO buffer 104 are analyzed.
The judgment history of voiced / non-voiced sound accumulated in the FIFO buffer 102 is continuously voiced over a certain number of blocks, and the number of zero-crossings accumulated in the FIFO buffer 104 as in the dial tone portion shown in FIG. When the difference history of is smaller than the threshold value continuously in the block, a response indicating that the input audible sound is the dial tone of the answering machine 1 is output to the control unit 7 as a determination result.

【0013】制御部7は留守番電話1の発信音である旨
の判定結果を識別条件判定部105から受信すると、音
声情報送出部6を制御してここから留守番電話1に対し
て予め指定してある音声情報を公衆網2を介し送出す
る。この発明は、留守番電話の応答メッセージに引続い
て一定時間連続して送出される発信音は周期的な単音或
は単音の組合わせにより構成される連続音であるので、
この発信音のブロックは一定時間連続して有音であると
共に、この発信音の同ブロックの零交差数の変動はその
他のブロックと比較して著しく小さい、という点に着目
するものである。ここで、可聴音の各ブロックが予め定
められたブロック数だけ連続して有音であると判断され
ると共に、各ブロックの零交差数の変動が予め定められ
たブロックだけ連続して一定値以下である場合、この連
続した可聴音のブロックを発信音のブロックであると判
定することにより、留守番電話において応答メッセージ
に引続いて送出される可聴音の内の発信音だけを検出す
ることができる。
When the control unit 7 receives the determination result indicating that it is the dial tone of the answering machine 1, from the identification condition determining unit 105, it controls the voice information sending unit 6 to pre-designate the answering machine 1 from here. A certain voice information is transmitted via the public network 2. According to the present invention, since the dial tone transmitted continuously for a certain period of time following the answering message of the answering machine is a continuous tone or a continuous tone composed of a combination of single tones,
It should be noted that the block of the dial tone is sounded continuously for a certain period of time, and the fluctuation of the number of zero crossings of the block of the dial tone is significantly smaller than that of the other blocks. Here, it is determined that each block of the audible sound is continuously voiced for a predetermined number of blocks, and the fluctuation of the number of zero crossings of each block is continuously for a predetermined block or less than a certain value. If it is, it is possible to detect only the dial tone of the audible tones that are sent subsequently to the response message in the answering machine by determining that this continuous block of the audible tone is the block of the dial tone. .

【0014】[0014]

【発明の効果】以上の通りであって、この発明の留守番
電話の発信音検出方法によれば、可聴音のブロック内平
均電力とブロック内零交差数を分析することにより留守
番電話から送出される可聴音の有音/無音を判定すると
共に、零交差数の時間変動を逐次分析して留守番電話の
発信音の検出を簡単確実に実施することができる。
As described above, according to the dial tone detection method of the answering machine of the present invention, the audible tone is transmitted from the answering machine by analyzing the average power in the block and the number of zero crossings in the block. The presence / absence of audible sound can be determined, and the time variation of the number of zero crossings can be sequentially analyzed to easily and reliably detect the dial tone of the answering machine.

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

【図1】この発明の実施例を説明するブロック図。FIG. 1 is a block diagram illustrating an embodiment of the present invention.

【図2】留守番電話の応答メッセージ波形、この波形の
有音/無音の判定結果、この波形の零交差数を示すグラ
フ。
FIG. 2 is a graph showing a response message waveform of an answering machine, a sound / no-sound determination result of this waveform, and the number of zero crossings of this waveform.

【符号の説明】[Explanation of symbols]

1 留守番電話 2 公衆電話回線網 3 着信検出部 4 発呼部 5 通話回路 6 音声情報送出部 7 制御部 100 発信音検出装置 101 有音/無音検出部 102 FIFOバッファ 103 零交差数計数部 104 FIFOバッファ 105 識別条件判定部 106 平均電力検出部 1 Answering machine 2 Public telephone line network 3 Incoming call detection section 4 Calling section 5 Calling circuit 6 Voice information sending section 7 Control section 100 Dial tone detection device 101 Voice / silence detection section 102 FIFO buffer 103 Zero crossing counting section 104 FIFO Buffer 105 Identification condition determination unit 106 Average power detection unit

Claims (1)

【特許請求の範囲】[Claims] 【請求項1】 留守番電話から応答メッセージに引き続
いて発信される発信音を検出する留守番電話の発信音検
出方法において、 留守番電話から送出される可聴音を複数の時系列ブロッ
クに分割し、 各時系列ブロックが有音であるか或は無音であるかを有
音/無音時系列データとして出力し、 各時系列ブロックの零交差数の変動を零交差数変動時系
列データとして出力し、 有音/無音時系列データおよび零交差数変動時系列デー
タのブロック毎の変動に基づいて、可聴音の時系列ブロ
ックが一定ブロック数に亘り連続して有音であると共
に、時系列ブロックの零交差数の変動が或る一定値より
も小である場合、その連続するブロックにおける可聴音
は留守番電話から発信される発信音であると判定する、 ことを特徴とする留守番電話の発信音検出方法。
1. A method of detecting a dial tone of an answering machine, which detects a dial tone transmitted from an answering machine following a response message, wherein an audible tone transmitted from the answering machine is divided into a plurality of time series blocks, and each time block is divided. Whether the sequence block is voiced or silent is output as voiced / non-voiced time series data, and the fluctuation of the zero-crossing number of each time-series block is output as zero-crossing number fluctuation time-series data. / Silence time series data and fluctuation of zero crossings Based on the fluctuation of each block of time series data, the time series blocks of audible sound are continuously voiced over a fixed number of blocks, and the number of zero crossings of time series blocks If the fluctuation of is less than a certain fixed value, it is determined that the audible sound in the consecutive blocks is the dial tone transmitted from the answering machine. Detection method.
JP5318482A 1993-12-17 1993-12-17 Dial tone detecting method for automatic answering telephone Pending JPH07177217A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP5318482A JPH07177217A (en) 1993-12-17 1993-12-17 Dial tone detecting method for automatic answering telephone

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP5318482A JPH07177217A (en) 1993-12-17 1993-12-17 Dial tone detecting method for automatic answering telephone

Publications (1)

Publication Number Publication Date
JPH07177217A true JPH07177217A (en) 1995-07-14

Family

ID=18099613

Family Applications (1)

Application Number Title Priority Date Filing Date
JP5318482A Pending JPH07177217A (en) 1993-12-17 1993-12-17 Dial tone detecting method for automatic answering telephone

Country Status (1)

Country Link
JP (1) JPH07177217A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111951834A (en) * 2020-08-18 2020-11-17 珠海声原智能科技有限公司 Method and device for detecting voice existence based on ultralow computational power of zero crossing rate calculation

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111951834A (en) * 2020-08-18 2020-11-17 珠海声原智能科技有限公司 Method and device for detecting voice existence based on ultralow computational power of zero crossing rate calculation

Similar Documents

Publication Publication Date Title
CA2569660C (en) Method and system for identifying a party answering a telephone call based on simultaneous activity
US6298055B1 (en) Early detection of in-band signals in a packet voice transmitter with reduced transmission delay
EP1224813B1 (en) Method and apparatus for early detection of dtmf signals in voice transmissions over an ip network
KR19980080615A (en) Voice activity detection method and apparatus
JPH07212480A (en) Device for sensing signaling of disconnection
US5428662A (en) Detecting make-break clicks on a telephone line
US5255340A (en) Method for detecting voice presence on a communication line
JPH05207526A (en) Sorting of signal of telephone and transmission method and system of message by telephone
JPH07177217A (en) Dial tone detecting method for automatic answering telephone
US5189669A (en) Cell transfer apparatus and method using a variable rate codec
JPH07104676B2 (en) Adaptive voicing end detection method
JPH0728486A (en) Voice compression device
US6046977A (en) Method and apparatus for detecting signal misalignment
JPH07110021B2 (en) Interactive voice response device
JP2804534B2 (en) Voice packet transmitting device and receiving device
JP3231699B2 (en) Voice detector, voice detection method, and high-efficiency terminal device
JPH0294743A (en) Dtmf signal receiver
JP3226068B2 (en) Data transmission equipment
JPH01306897A (en) Speech detection circuit
KR0184463B1 (en) Self-restoring apparatus of speech path of telephone terminal unit
JP2556978B2 (en) Interactive answering machine
JPH0738661B2 (en) Automatic voice return telephone
JPH0621900A (en) Silence suppression variable controller for voice signal
JPH0573085A (en) Voicelessness detection device and encoding device
JPH03276947A (en) Interactive automatic answering telephone set