JPH0535294A - Speech recognition device - Google Patents

Abstract

PURPOSE:To prevent an erroneous control by an external device caused by an erroneous recognition even though it is operated in a noisy place. CONSTITUTION:A voice detecting section 13 sets the minimum voice level in accordance with the noise signal level just prior to the start of a voice detection and the level is stored in a level storage circuit 32. And when a voice is detected, a level comparator circuit 33 compares whether the maximum level of that voice is larger than the minimum voice level or not and only if it is large, a degree of similarity decision section 23 decides the degree of similarity with a voice standard pattern. By this arrangement, an erroneous control by an external device caused by an erroneous recognition is prevented even though the noise is large.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a voice recognition device capable of controlling an external device by using a voice, and is liable to erroneously recognize a voice particularly under a situation where noise or noise is large. The present invention relates to a voice recognition device capable of preventing an error in control of an external device in a situation.

[0002]

2. Description of the Related Art Various voice recognition devices for recognizing voice and controlling external devices have been put into practical use. In such a device, if all sounds or voices are to be recognized, there is a problem that they are easily affected by noise or the like. Therefore, a certain type of voice recognition device has heretofore adopted a method in which a threshold for detecting voice is set high and voice recognition is performed only when a large voice is input to control an external device. Further, in other voice recognition devices, while setting the threshold for voice detection to a normal level, the degree of similarity in voice recognition is severely determined until words such as predetermined command words are recognized. The method of controlling the external device was adopted to reduce the false recognition.

[0003]

However, in the former one of the conventional voice recognition devices, the external device operates when a large noise exceeding the threshold for voice detection is input. was there. Also, in the latter device, since the voice level is not a problem, noise and conversations that occur around the device may be mistakenly recognized as command words.In such a case, the user may not control the external device. There is a disadvantage that the external device may malfunction even when it is not desired.

Therefore, an object of the present invention is to provide a voice recognition device which can prevent erroneous control of an external device due to erroneous recognition.

[0005]

SUMMARY OF THE INVENTION According to the present invention, a voice detection unit for detecting voice and a minimum voice level set in conjunction with a noise signal level immediately before the voice detection unit starts voice detection are set. A voice level setting means, a level comparing means for comparing whether or not the maximum level of the voice is higher than the minimum voice level when the voice detecting section finishes detecting the voice, and a maximum level of the voice is the minimum voice level. Similarity determination means for determining the similarity between the voice and a pre-registered voice standard pattern when it is determined to be greater than the level,
A voice recognition device is provided with a recognition result output unit that outputs a determination result of similarity as a voice recognition result.

That is, according to the first aspect of the invention, the voice detection unit sets the minimum voice level in association with the noise signal level immediately before the voice detection is started, and when the voice is detected, the maximum level in the voice is set to the minimum. Only when the voice level is higher than the voice level, the similarity with the voice standard pattern is determined so that erroneous control of an external device due to erroneous recognition can be prevented.

[0007]

EXAMPLES The present invention will be described in detail below with reference to examples.

FIG. 1 shows an outline of the configuration of a voice recognition device according to an embodiment of the present invention. A voice signal 12 output from a microphone 11 for capturing the voice of the speaker
Is input to the voice detection unit 13. The level monitoring unit 14
The input signal level of the audio signal 12 is constantly monitored. The detected voice signal 15 is input to the voice analysis unit 16 and encoded to extract a voice pattern 17. The extracted voice pattern 17 is input to the similarity calculation unit 18. The similarity calculation unit 18 uses the standard pattern 2 stored in the storage unit 19.
1 is used to calculate the similarity with the voice pattern 17. Here, the standard pattern 21 is a coded number or the like in addition to the command voice.

The result 22 regarding the similarity calculated by the similarity calculating unit 18 is sent to the similarity determining unit 23. In the similarity determination unit 23, the voice pattern 17
If the degree of similarity is high, it is determined that the voice of the speaker matches the standard pattern 21. In this case, the recognition result 24 is sent to the recognition result sending unit 25. The recognition result sending unit 25 outputs a control signal 26 for controlling an external device (not shown) based on this. The level monitoring unit 14, the voice analysis unit 16, the similarity calculation unit 18, the storage unit 19, the similarity determination unit 23, and the recognition result transmission unit 25 described above are each connected to the control unit 27, and control is performed by this. It is supposed to be.

FIG. 2 specifically shows the level monitoring unit shown in FIG. The level monitoring unit 14 includes a level monitoring circuit 31 connected to the voice detection unit 13, a level storage circuit 32, and a level comparison circuit 33. The level monitoring circuit 31 constantly monitors the input signal level of the voice detection unit 13. This level monitoring circuit 3
When the voice detection is started, the level storage circuit 32 connected to 1 sets the minimum voice level linked to the level of the noise signal of this voice detection and stores it. The level comparison circuit 33 compares the audio signal level detected by the level monitoring circuit 31 during audio detection with the minimum audio level stored in the level storage circuit 32. Then, at the end of voice detection, when the maximum level of the voice signal during voice detection exceeds the minimum voice level, a signal for causing the control unit 27 to perform similarity determination is transmitted. . In addition,
The level monitoring unit 14 is controlled by a CPU (central processing unit) (not shown) together with other circuit parts of the voice recognition device shown in FIG. A program for this control is also stored in a ROM (read only memory) not shown.

The operation of the speech recognition apparatus configured as described above will be specifically described with reference to FIG. It is assumed that the standard pattern of the control voice stored in the storage unit 19 of the voice recognition device shown in FIG. 1 is "Kido". When the control operation of the voice recognition is started, as shown in FIG. 3, the voice detection waiting state is entered (step S101). In this state, the voice detection unit 13 constantly monitors the signal from the microphone 11. When voice detection is started (Y), the level monitoring circuit 31 shown in FIG. 2 monitors the signal level of the voice analysis unit 16 and sets the minimum voice level linked to the level of the noise signal immediately before the start of voice detection. (Step S102). That is, when the speaker utters "○○○" and the temporally leading end portion of the voice is detected, the minimum voice level is set at the same time. This minimum voice level is stored in the level storage circuit 32 of FIG.

When the minimum voice level is set, the level monitor circuit 31 monitors the voice level, and the level comparison circuit 33 stores the voice signal level detected by the level monitor circuit 31 during voice detection and the level storage circuit 32. Compare with the stored minimum voice level. At the same time, the similarity calculation unit 18 calculates the similarity between the standard pattern stored in the storage unit 19 and the input voice pattern (step S10).
3).

The monitoring of the voice level and the calculation of the similarity between the voice patterns are performed until the voice detection is completed (step S104). When the voice detection is completed (Y), the level comparison circuit 33 determines whether or not the maximum voice level of "○○○" uttered by the speaker is higher than the minimum voice level described above (step S105). As a result, when it is determined that the level is equal to or lower than the minimum voice level (N), the similarity determination unit 23 does not perform the similarity determination, and the process returns to step S101 and the next voice is reproduced. Waiting for input.

On the other hand, "○○○" uttered by the speaker
When the maximum level of the voice in is higher than the minimum voice level (step S105; Y), the similarity determination unit 23 determines the similarity. Then, if it is determined that the speaker uttered a voice that is not similar to the registered standard pattern (step S106; N), step S10.
The process returns to 1 and waits for the next voice input.

On the other hand, if a similarity higher than a preset similarity determination value is obtained, such as when the speaker utters "Kido", it is determined that the similarity is similar to the standard pattern. (Step S106; Y). in this case,
Upon receiving the recognition result 24, the recognition result sending unit 25 outputs the control signal 26 corresponding to the recognition result to the external device (step S107). This completes a series of operations (end).

[0016]

As described above, according to the present invention, the minimum voice level is set in association with the noise level immediately before the start of voice detection, and the voice standard pattern is set only when the voice signal exceeding the minimum voice level is input. Since it was decided to determine the similarity of the, it is possible to prevent erroneous recognition due to erroneous detection of voice even in a noisy environment or when noise is generated due to conversation in the surroundings. There is an effect that the control can be surely prevented.

[Brief description of drawings]

FIG. 1 is a block diagram showing an outline of a configuration of a voice recognition device according to an embodiment of the present invention.

FIG. 2 is a block diagram specifically showing a level monitoring unit shown in FIG.

FIG. 3 is a flowchart showing how control is performed by the voice recognition device according to the present embodiment.

[Explanation of symbols]

11 microphones 13 Voice detector 14 Level monitoring unit 16 Speech analysis section 18 Similarity calculation section 19 memory 23 Similarity Determining Section 25 Recognition result sending unit 27 Control unit 31 Level monitoring circuit 32 level memory circuit 33 level comparison circuit

Claims (2)

[Claims] 1. A voice detecting section for detecting voice, a minimum voice level setting means for setting a minimum voice level in association with a noise signal level immediately before the voice detecting section starts detecting voice, and the voice When the detection unit finishes detecting the voice, level comparison means for comparing whether or not the maximum level of the voice is higher than the minimum voice level, and it is determined that the maximum level of the voice is higher than the minimum voice level. And a recognition result output unit for outputting a similarity determination result as a voice recognition result. The similarity determination unit determines a similarity between the voice and a standard voice pattern registered in advance. Voice recognition device. 2. A voice analysis unit that analyzes and encodes a voice, and a storage unit that encodes a voice to be recognized by a device and stores the voice as a standard voice pattern, and the similarity determination unit stores in this storage unit. 2. The voice recognition device according to claim 1, wherein the stored voice standard pattern is compared with the voice encoded by the voice analysis unit to determine the degree of similarity.