KR890001026Y1 - Device for speech recognition - Google Patents
Device for speech recognition Download PDFInfo
- Publication number
- KR890001026Y1 KR890001026Y1 KR2019850018380U KR850018380U KR890001026Y1 KR 890001026 Y1 KR890001026 Y1 KR 890001026Y1 KR 2019850018380 U KR2019850018380 U KR 2019850018380U KR 850018380 U KR850018380 U KR 850018380U KR 890001026 Y1 KR890001026 Y1 KR 890001026Y1
- Authority
- KR
- South Korea
- Prior art keywords
- speech recognition
- counter
- voice
- waveform
- multiplexer
- Prior art date
Links
- 238000010586 diagram Methods 0.000 description 4
- 239000003990 capacitor Substances 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 230000000875 corresponding effect Effects 0.000 description 1
- 230000005236 sound signal Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/78—Detection of presence or absence of voice signals
Landscapes
- Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
내용 없음.No content.
Description
제1a도는 종래 음성 인식장치의 기본 구성도.Figure 1a is a basic configuration of the conventional speech recognition device.
제1b도는 종래 음성인식 장치의 실시 회로도.1B is an implementation circuit diagram of a conventional voice recognition device.
제2도는 본 고안 음성 인식 시스템의 구성도.2 is a block diagram of the present invention speech recognition system.
제3도는 본 고안 음성 인식 시스템중 기울기 변화 계수기의 상세 회로도.3 is a detailed circuit diagram of a gradient change counter of the present invention speech recognition system.
제4도는 본 고안 음성 인식 시스템의 작동을 설명하기 위한 파형도.4 is a waveform diagram for explaining the operation of the present invention speech recognition system.
본 고안은 음성마다의 특유 주파수 파형을 인식하기 위한 시스템에 관한 것으로서 특히 시간 도메인(Time Domain)에서 빠른 시간내에 음성의 특징을 정확하게 추출할 수 있는 음성 인식 시스템에 관한 것이다.The present invention relates to a system for recognizing a specific frequency waveform for each voice, and more particularly, to a voice recognition system capable of accurately extracting voice features in a time domain.
종래에는 제1a도에서와 같이 소정의 음성 입력(30)을 프리앰파시스회로(31) 및 필터뱅크(32)와 0교차 계수기(33)를 통하여 멀티 플렉서(34)에 인가한 후 아날로그 디지탈 변환기(35)에서 디지탈신호로 변화되어 주컴퓨터(36)에 인가하는 구성으로서 특히 핵심이되는 0교차 계수기(33)의 구성은 제1도 b와 같이 비교기(331)와 정전압 다이오드(ZD) 및 계수기(332)로 구성되어 있어 음성 신호 파형이 0축을 교차하는 점의 갯수를 계수하도록 되어 있음을 알 수 있다. 그러나 이러한 구성으로는 제4도 e, f에서의 서로 다른 음성 파형을 단지 0축을 교차하는 점의 갯수로 판단한다면 종래의 장치로서는 동일한 음성으로 판단하기 때문에 실제적으로 활용하기에는 문제점이 많다.Conventionally, as shown in FIG. 1A, a predetermined voice input 30 is applied to the multiplexer 34 through the pre-emphasis circuit 31, the filter bank 32, and the zero crossing counter 33, and then analog digital. The configuration of the zero crossing counter 33, which is a key component, which is converted into a digital signal by the converter 35 and applied to the main computer 36, is composed of a comparator 331, a constant voltage diode ZD, and the like as shown in FIG. It can be seen that the counter 332 is configured to count the number of points where the audio signal waveform intersects the zero axis. However, in such a configuration, if the different voice waveforms in Figs. 4E and F are judged only by the number of points crossing the 0 axis, the conventional devices are judged to be the same voice, so there are many problems in practical use.
본 발명은 이러한 종래의 단점을 없이하도록 음성신호 파형의 피크점과 계곡점을 구분하여 계수하여 음성을 확연히 구분 인지할 수 있게 되도록 할 수 있는 음성 인식 시스템을 제공하는 것을 목적으로 하며 이하 첨부된 도면을 참조하면서 본 발명을 상세히 설명하면 다음과 같다.An object of the present invention is to provide a speech recognition system capable of distinguishing and recognizing speech by distinguishing and counting peak and valley points of a speech signal waveform so as to eliminate these disadvantages. The present invention will be described in detail with reference to the following.
우선 제2도를 참조하면 음성입력단자(1)가 연결된 프리엠파시스부(2), 필터뱅크(3), 0교차 계수기(4), 멀티플렉서(5), 아날로그/디지탈 변환기(6), 및 주컴퓨터(7)로 구성된 음성인식장치에서, 프리엠파시스부(2)와 멀티 플렉서(5) 사이에 기울기 변화 계수기(8)를 설치하되, 기울기 변화 계수기(8)는 미분기(81), 0교차 검출기(82)및 계수기(83)으로 구성된다.First, referring to FIG. 2, the pre-emphasis unit 2 to which the voice input terminal 1 is connected, the filter bank 3, the zero crossing counter 4, the multiplexer 5, the analog / digital converter 6, and In the speech recognition device composed of the main computer 7, a gradient change counter 8 is provided between the pre-emphasis unit 2 and the multiplexer 5, and the gradient change counter 8 is a differentiator 81, It consists of a zero crossing detector 82 and a counter 83.
이러한 본 발명의 작동을 설명하자면 제4도 e, f와 같은 음성신호를 미분하여 미분파형의 0교차점을 검출하면 음성신호의 피크점과 계곡점이 되는 것이다. 이러한 고정은 다음식과 같이 표현할수 있다.To explain the operation of the present invention, when the zero crossing point of the differential waveform is detected by differentiating the voice signal as shown in FIG. 4 e and f, the peak point and valley point of the voice signal are obtained. This fixation can be expressed as
그렇치 않은 경우 즉 m〈0, 또는 m〉n-1이면 w(m)=0가 된다.Otherwise, m <0, or m> n-1, w (m) = 0.
윗식의 조건은 제3도에서 만족된다.The condition of the above expression is satisfied in FIG.
제3도를 참조하면 저항(R1)과 콘덴서(C1)연산증폭기(OP1)및 연산증폭기(OP1)의 궤환부(FB)로 구성된 미분기(81)에서 제4도 a와 같은 입력신호를 미분하여 제4c도와 같은 파형이 발생되어 제4도 c의 미분파형을 제4도 a의 파형에 비교하여 보면 제4a도의 파형에서 피크점과 계곡점 대응하는 부위가 모두 0레벨에서 교차되고 있음을 알 수 있다. 이러한 신호는 이후 다시 비교기(CO)로 구성된 0교차 검출기 (82)에 입력되어 제4도 d와 같은 펄스 파형이 출력된다. 이 파형은 제너다이오드 (ZD3)로서 상측북 컬트되어 계수기(83)에서 계속된다.Referring to FIG. 3, the input shown in FIG. 4A is different from the differentiator 81 composed of the resistor R 1 , the capacitor C 1 , the operational amplifier OP 1 , and the feedback part FB of the operational amplifier OP 1 . Differentiate the signal to generate a waveform like Figure 4c. Compare the derivative waveform of Figure 4c with the waveform of Figure 4a. In the waveform of Figure 4a, all the points corresponding to the peak and valley points cross at zero level. It can be seen that. This signal is then input to the zero crossing detector 82, which is composed of the comparator CO, to output a pulse waveform as shown in FIG. This waveform is cultivated upward as zener diode ZD 3 and continues at counter 83.
그러므로서 본 발명에서는 피크점과 계곡점이 검출 계수되는 것이고 이 계수 출력이 0교차 계수기(4)의 출력과 상호 연관되어 보다 정확한 음성을 인식할 수 있는 것이다.Therefore, in the present invention, the peak point and the valley point are detected and the coefficient output is correlated with the output of the zero crossing counter 4 so that a more accurate voice can be recognized.
이상에서 설명된 바와 같이 본 발명에 의하면 음성을 0교차점 및 피크점과 계곡점으로 구분하여 인식하게 되므로 음성 인식을 정확히 할 수 있는 것이다.As described above, according to the present invention, since the voice is recognized by being divided into 0 crossing points and peak points and valley points, the voice recognition can be accurately performed.
Claims (2)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| KR2019850018380U KR890001026Y1 (en) | 1985-12-31 | 1985-12-31 | Device for speech recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| KR2019850018380U KR890001026Y1 (en) | 1985-12-31 | 1985-12-31 | Device for speech recognition |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| KR870011017U KR870011017U (en) | 1987-07-15 |
| KR890001026Y1 true KR890001026Y1 (en) | 1989-03-28 |
Family
ID=19247827
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| KR2019850018380U KR890001026Y1 (en) | 1985-12-31 | 1985-12-31 | Device for speech recognition |
Country Status (1)
| Country | Link |
|---|---|
| KR (1) | KR890001026Y1 (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR100452109B1 (en) * | 2002-10-07 | 2004-10-12 | 주식회사 아큐죤 | Tracking device and method of voice signal |
-
1985
- 1985-12-31 KR KR2019850018380U patent/KR890001026Y1/en not_active IP Right Cessation
Also Published As
| Publication number | Publication date |
|---|---|
| KR870011017U (en) | 1987-07-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CA1177588A (en) | Digital circuit and method for the detection of call progress tones in telephone systems | |
| US3688126A (en) | Sound-operated, yes-no responsive switch | |
| US4039754A (en) | Speech analyzer | |
| US4216463A (en) | Programmable digital tone detector | |
| JP2609752B2 (en) | Voice / in-band data identification device | |
| US3335225A (en) | Formant period tracker | |
| GB2031696A (en) | Electronic speech detector | |
| US4534041A (en) | Digital circuit for determining the envelope frequency of PCM encoded call progress tones in a telephone system | |
| KR890001026Y1 (en) | Device for speech recognition | |
| EP0323871A3 (en) | Voltage to frequency converter and its application in a fibre optic transmission system | |
| US4160922A (en) | Method of generating a direct current control signal from a noisy alternating current signal | |
| GB981383A (en) | Sound analyzing system | |
| US4786797A (en) | Electro-optical data handling system with noise immunity | |
| US3225141A (en) | Sound analyzing system | |
| KR900010403A (en) | Ringing Detector and Ringing Signal Detection Method | |
| US4022990A (en) | Technique and apparatus for measuring the value of a capacitance in an electrical circuit such as a telephone communication line | |
| US4009353A (en) | Analog signal detection | |
| SU1363275A1 (en) | Apparatus for identifying information | |
| KR100198057B1 (en) | Method and apparatus for extracting the property of speech signal | |
| SU1370460A1 (en) | Voltage instrument transducer of variable relactance pickups | |
| EP0566928A2 (en) | Tone detection method for telephone subset | |
| JPS6126089A (en) | Musical scale detector | |
| SU454493A1 (en) | Pulse amplitude detection device | |
| JPS6137035Y2 (en) | ||
| US20050253628A1 (en) | Current-voltage conversion circuit |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| A201 | Request for examination | ||
| E701 | Decision to grant or registration of patent right | ||
| REGI | Registration of establishment | ||
| FPAY | Annual fee payment |
Payment date: 19951226 Year of fee payment: 8 |
|
| LAPS | Lapse due to unpaid annual fee |