JPH036518B2 - - Google Patents

Info

Publication number
JPH036518B2
JPH036518B2 JP57002675A JP267582A JPH036518B2 JP H036518 B2 JPH036518 B2 JP H036518B2 JP 57002675 A JP57002675 A JP 57002675A JP 267582 A JP267582 A JP 267582A JP H036518 B2 JPH036518 B2 JP H036518B2
Authority
JP
Japan
Prior art keywords
plosive
oral airflow
oral
airflow
sound
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 - Lifetime
Application number
JP57002675A
Other languages
Japanese (ja)
Other versions
JPS58121098A (en
Inventor
Takeo Murata
Nobuhisa Kadowaki
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.)
National Institute of Advanced Industrial Science and Technology AIST
Original Assignee
Agency of Industrial Science and Technology
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 Agency of Industrial Science and Technology filed Critical Agency of Industrial Science and Technology
Priority to JP267582A priority Critical patent/JPS58121098A/en
Publication of JPS58121098A publication Critical patent/JPS58121098A/en
Publication of JPH036518B2 publication Critical patent/JPH036518B2/ja
Granted legal-status Critical Current

Links

Description

【発明の詳細な説明】 本発明は破裂音を検出する破裂音検出装置に関
するものである。
DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a plosive sound detection device for detecting plosive sounds.

従来、マイクロホンより通常の方法でピツクア
ツプした音声の特徴から破裂音を確実に抽出する
方法は知られていない。
Conventionally, there is no known method for reliably extracting plosive sounds from the characteristics of sounds picked up by a microphone using a normal method.

本発明は発話にともなう口気流情報より、破裂
音を確実、簡便かつリアルタイムで抽出する破裂
音検出装置を提供することを目的とする。
An object of the present invention is to provide a plosive sound detection device that reliably, simply, and in real time extracts plosive sounds from oral airflow information associated with speech.

以下、本発明の実施例について図面とともに説
明する。第1図は、本発明の破裂音検出装置の実
施例を示すブロツク図である。
Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing an embodiment of the plosive sound detection device of the present invention.

図において、1,2及び3は、発話時の口気流
の流速および流速の変化(加速度)を検出する口
気流検出器であり、たとえば、熱線式風速検出器
や風防を除去したマイクロホンなどで構成され
る。これらの口気流検出機1,2,3は第2図に
示すように発話者の口の前方で発話にともなう口
気流が検出できる距離において、それぞれ破裂音
p発話時の最大口気流流速の位置とt発話時の最
大口気流流速の位置とt発話時の最大口気流流速
の位置に配置される。4,5及び6は夫々口気流
検出器1,2及び3の出力信号が破裂音の特徴を
有するか否かを判定する判定回路であり、たとえ
ば最も簡単には口気流検出器の出力信号の大きさ
が予め定められた閾値をこえるかどうかを弁別す
る電圧比較回路から構成される。
In the figure, 1, 2, and 3 are oral airflow detectors that detect the flow velocity of the oral airflow and changes in the flow velocity (acceleration) during speech, and are composed of, for example, a hot wire wind velocity detector or a microphone with a windshield removed. be done. As shown in Fig. 2, these mouth airflow detectors 1, 2, and 3 detect the position of the maximum oral airflow velocity when the plosive p is uttered, respectively, at a distance in front of the speaker's mouth where the oral airflow accompanying speech can be detected. and the position of the maximum oral airflow velocity at the time of t speech, and the position of the maximum oral airflow velocity at the time of t speech. Numerals 4, 5 and 6 are determination circuits for determining whether or not the output signals of the oral air flow detectors 1, 2 and 3 have the characteristics of a plosive. It consists of a voltage comparison circuit that discriminates whether the magnitude exceeds a predetermined threshold.

7は、判定回路4,5,6のいずれかが破裂音
であるという信号を出力したときに破裂音信号を
出力するOR回路であり、OR回路出力から破裂
音であるか否かの判定結果を得ることができる。
7 is an OR circuit that outputs a plosive sound signal when any of the determination circuits 4, 5, and 6 outputs a signal indicating that it is a plosive sound, and the result of determining whether or not it is a plosive sound is determined from the output of the OR circuit. can be obtained.

実験によれば、破裂音発話時の口気流は、非破
裂音発話時の口気流に対して著しい特徴を有する
ことが明かになつた。たとえば、破裂音発話時の
口気流の流速波形は破裂音の種類に特有な方向に
おいて急激な立上りを有するパルス状となる。こ
れに対して非破裂音発話時の口気流の流速波形
は、非常に小さいか又は、緩除な立上りを有する
波形となる。第3図イは、これらの様子を示すシ
エーマであり、aは破裂音、b及びcは非破裂音
発話時の口気流流速波形を示す。第3図ロは、破
裂音a′及び非破裂音(b′及びc′)発話時の口気流
の加速度を示す図であり、加速度を検出する口気
流検出器の出力又は口気流流速を検出する口気流
検出器の出力を電気回路的に微分することによつ
てえられる。第3図ロから明らかな様に、破裂音
に対しては、振巾の大きいパルス状の波形にな
り、非破裂音に対しては、振巾の小さい波形が得
られる。
Experiments have revealed that the oral airflow during plosive utterances has remarkable characteristics compared to the oral airflow during non-plosive utterances. For example, the flow velocity waveform of the oral airflow when a plosive is uttered has a pulse shape with a sharp rise in a direction specific to the type of plosive. On the other hand, the flow velocity waveform of the oral airflow during non-plosive speech is very small or has a gradual rise. FIG. 3A is a schema showing these situations, where a shows the plosive sound, and b and c show the oral airflow velocity waveforms when a non-plosive sound is uttered. Figure 3B is a diagram showing the acceleration of the oral airflow when uttering the plosive a' and the non-plosive sounds (b' and c'), and detects the output of the oral airflow detector that detects the acceleration or the oral airflow velocity. It is obtained by differentiating the output of the oral airflow detector using an electrical circuit. As is clear from FIG. 3B, a pulse-like waveform with a large amplitude is obtained for a plosive sound, and a waveform with a small amplitude is obtained for a non-plosive sound.

さらに実験によれば、上記のように破裂音の特
有の口気流信号は破裂音の種類によつて定まる特
定位置に口気流検出器を置いた時に得られること
がわかつた。
Furthermore, experiments have shown that the oral airflow signal unique to plosives can be obtained when the oral airflow detector is placed at a specific position determined by the type of plosive.

第2図bは実験による結果例を示すもので、特
定の発話者によつて破裂音|pa|,|ta|,|ka
|又は|ba|,|da|,|ga|を一定に発話した
場合の口気流の分布を測定したものである。ここ
で、,およびは、発音点ロを0とした
ときの各破裂音|p|,|b|,|k|,|g|お
よび|t|,|d|の口気流分布の中心軸を示す
もので、水平軸からの角度∠XOA,∠XOB
および∠XOCは、発話者によつて異なるが、そ
れぞれほぼ0゜〜20゜、30゜〜50゜および50゜〜70゜と
な
る。したがつて、口気流検出は、口気流検出器を
これらのほぼ中心軸上に置くことによつて可能に
なる。
Figure 2b shows an example of the experimental results.
The distribution of oral airflow was measured when | or |ba|, |da|, and |ga| were uttered at a constant rate. Here, , and are the central axes of the oral airflow distribution for each plosive |p|, |b|, |k|, |g|, and |t|, |d| when the pronunciation point B is set to 0. The angles ∠XOA and ∠XOB from the horizontal axis are
and ∠XOC vary depending on the speaker, but are approximately 0° to 20°, 30° to 50°, and 50° to 70°, respectively. Oral airflow detection is therefore possible by placing the oral airflow detector approximately on these central axes.

具体的には、第2図aに示すように、破裂音|
p|,|b|に対しては10に示す位置に、|k
|,|g|に対しては20の位置に、|t|,|d
|に対しては30の位置に口気流検出器を配置す
る。もし口気流検出器の設置位置が上記位置より
はずれると、破裂音発話時に口気流検出器から得
られる出力波形が第3図a又はa′の様な特徴的な
波形でなくなり、非破裂音との区別が困難、或い
は不正確となる。口気流検出器1,2,3として
前述の口気流の加速度を検出するものを用いる
か、或いは口気流検出を微分回路を通す場合に
は、検出の精度を向上させるとともに、口気流検
出器の配置に若干余裕を与えることができる。
Specifically, as shown in Figure 2 a, the plosive sound |
For p|, |b|, |k
|t|, |d at position 20 for |, |g|
For |, an oral airflow detector is placed at position 30. If the installation position of the oral airflow detector is deviated from the above-mentioned position, the output waveform obtained from the oral airflow detector during plosive utterances will no longer have the characteristic waveform shown in Figure 3 a or a', and will become a non-plosive sound. It is difficult or inaccurate to distinguish between If the aforementioned oral airflow detectors 1, 2, and 3 are used to detect the acceleration of the oral airflow, or if the oral airflow detection is passed through a differential circuit, the accuracy of detection is improved and the This allows for some leeway in placement.

それぞれの破裂音に対する口気流検出器の出力
関係を第4図に示す。ここでは口気流検出器とし
て口気流の加速度を検出するものか、或いは微分
回路を含む口気流の速度を検出するものを用いた
場合について示している。同図から明らかなよう
に、破裂音|p|,|b|は口気流検出器1から、
|k|,|g|は2から、|t|,|d|は3から
それぞれ検出され、対応する判定回路4,5,6
で判定されることになる。いずれの判定回路から
も同図のような結果が得られなければ、OR回路
7から出力が生じないので非破裂音と判定する。
FIG. 4 shows the output relationship of the oral airflow detector for each plosive sound. Here, a case is shown in which a device that detects the acceleration of the oral airflow or a device that detects the velocity of the oral airflow that includes a differential circuit is used as the oral airflow detector. As is clear from the figure, the plosive sounds |p|, |b| are detected by the oral airflow detector 1.
|k|, |g| are detected from 2, |t|, |d| are detected from 3, and the corresponding judgment circuits 4, 5, 6
will be judged. If the results shown in the figure are not obtained from any of the determination circuits, no output is generated from the OR circuit 7, and the sound is determined to be a non-plosive sound.

以上の説明においては3個の口気流検出器を用
いたが、使用中における口気流検出器と口との相
対位置のずれ等を考慮する必要のある場合には、
3個以上の複数個の口気流検出器を用いて同じ考
えに基き破裂音検出装置を構成することができ
る。
In the above explanation, three oral airflow detectors were used, but if it is necessary to take into account the relative positional deviation between the oral airflow detector and the mouth during use,
A plosive sound detection device can be constructed based on the same idea using three or more oral airflow detectors.

本発明による破裂音検出装置では、複数の口気
流検出器を、例えば第2図aの10,20および
30のように異なる位置に設置される。このよう
な構成において、任意の破裂音を含む言葉を発話
したとき複数の口気流検出器のいずれかが破裂音
に特有の口気流を検出し特有の信号を出力する。
この破裂音特有の信号は判定回路4,5又は6の
いずれかにより破裂音と判定され、OR回路7か
らは破裂音信号が出力される。
In the plosive sound detection device according to the present invention, a plurality of oral airflow detectors are installed at different positions, for example, 10, 20 and 30 in FIG. 2a. In such a configuration, when a word containing an arbitrary plosive is uttered, one of the plurality of mouth air flow detectors detects the mouth air flow specific to the plosive and outputs a specific signal.
This plosive-specific signal is determined to be a plosive by one of the determination circuits 4, 5, or 6, and the OR circuit 7 outputs a plosive signal.

以上から本発明によれば発話にともなう口気流
情報より破裂音を確実、簡便かつリアルタイムで
抽出できる。
As described above, according to the present invention, plosive sounds can be extracted reliably, easily, and in real time from oral airflow information accompanying speech.

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

第1図は本発明による破裂音検出装置の実施例
を示すブロツク図、第2図aは第1図における口
気流検出器の配置を示す図、第2図bは破裂音の
分布を示す図、第3図は第1図実施例の動作を説
明するための波形図、第4図は破裂音と第1図に
おける口気流検出器の出力との関係を示す図であ
る。 1,2,3…口気流検出器、4,5,6…判定
回路、7…OR回路。
FIG. 1 is a block diagram showing an embodiment of the plosive sound detection device according to the present invention, FIG. 2 a is a diagram showing the arrangement of the mouth airflow detector in FIG. 1, and FIG. 2 b is a diagram showing the distribution of plosive sounds. 3 is a waveform diagram for explaining the operation of the embodiment shown in FIG. 1, and FIG. 4 is a diagram showing the relationship between a plosive sound and the output of the oral air flow detector shown in FIG. 1, 2, 3... Oral air flow detector, 4, 5, 6... Judgment circuit, 7... OR circuit.

Claims (1)

【特許請求の範囲】[Claims] 1 口の前方で破裂音p発話時の最大口気流流速
の位置とk発話時の最大口気流流速の位置とt発
話時の最大口気流流速の位置にそれぞれ設けた3
個の口気流検出器と、前記3個の口気流検出器の
各出力信号が破裂音の特徴を有するか否かを判定
する3個の判定回路と、前記3個の判定回路の出
力信号の和を出力するOR回路から構成され、破
裂音発話時の口気流の特徴から破裂音を検出する
ことを特徴とする破裂音検出装置。
1 At the front of the mouth, 3 were installed at the position of the maximum oral airflow velocity during the plosive p utterance, the maximum oral airflow velocity during the k utterance, and the maximum oral airflow velocity during the t utterance.
three determination circuits for determining whether or not each output signal of the three oral air flow detectors has the characteristics of a plosive, and a A plosive sound detection device is comprised of an OR circuit that outputs a sum, and is characterized in that it detects plosive sounds from the characteristics of oral airflow when uttering a plosive sound.
JP267582A 1982-01-13 1982-01-13 Plosive detector Granted JPS58121098A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP267582A JPS58121098A (en) 1982-01-13 1982-01-13 Plosive detector

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP267582A JPS58121098A (en) 1982-01-13 1982-01-13 Plosive detector

Publications (2)

Publication Number Publication Date
JPS58121098A JPS58121098A (en) 1983-07-19
JPH036518B2 true JPH036518B2 (en) 1991-01-30

Family

ID=11535877

Family Applications (1)

Application Number Title Priority Date Filing Date
JP267582A Granted JPS58121098A (en) 1982-01-13 1982-01-13 Plosive detector

Country Status (1)

Country Link
JP (1) JPS58121098A (en)

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS5923834B2 (en) * 1973-05-14 1984-06-05 三菱電機株式会社 2-needle sewing device

Also Published As

Publication number Publication date
JPS58121098A (en) 1983-07-19

Similar Documents

Publication Publication Date Title
US7308105B2 (en) Environmental noise monitoring
JPS6129000B2 (en)
US6311156B1 (en) Apparatus for determining aerodynamic wind of utterance
JPS58121098A (en) Plosive detector
US3846586A (en) Single oral input real time analyzer with written print-out
JPS5949742A (en) Apparatus for detecting exhalation force
JPS6329759B2 (en)
JPH036519B2 (en)
JPH036520B2 (en)
JPS6329760B2 (en)
JPS6258519B2 (en)
JPH0682275B2 (en) Voice recognizer
JPS63226691A (en) Standard pattern creation method
JPS6260719B2 (en)
JPH0412478B2 (en)
JPH0438612Y2 (en)
JPH0415480B2 (en)
JPH04211299A (en) Monosyllabic speech recognizer
JPH02297600A (en) consonant display device
JPH0640274B2 (en) Voice recognizer
JPH024918B2 (en)
JPS6260720B2 (en)
JPS6260718B2 (en)
JPS58120299A (en) H series sound detector
JPS6230640B2 (en)