JPH02134697A - voice recognition device - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of Industrial Application The present invention relates to a speech recognition device that can be used in noisy environments.

2. Description of the Related Art Conventionally, when performing voice recognition, as shown in FIG. was taking.

Problems to be Solved by the Invention However, in the conventional configuration, ambient noise is simultaneously picked up by the microphone along with the speaker's voice, so the problem is that that single sound can lead to incorrect interpretation of the audio cutout section, resulting in frequent erroneous recognition. It became.

SUMMARY OF THE INVENTION In order to solve the above problems, it is an object of the present invention to provide a speech recognition device that can appropriately determine a segment of speech.

Means for Solving the Problems In order to solve the above problems, the speech recognition device of the present invention includes a first microphone that picks up the speaker's voice, a second microphone installed in the seat, and a seat-mounted microphone. By the pressure sensor group installed at i and the signal from the pressure sensor group,
The configuration includes a signal selection unit that selects an input signal from the second microphone group, and a voice recognition unit that controls a segment of the input audio signal by the first microphone based on the signal from the signal selection unit. .

Operation With the above configuration, the signal selection unit selects the input signal with a large voice component of the speaker from the input voice from the second microphone group based on the signal from the pressure sensor group installed on the seat, and the voice recognition means selects the input signal that has a large voice component of the speaker. The selected signal accurately controls the audio cutout of the input audio signal from the first microphone.

Embodiment FIG. 1 shows a speech recognition device in an embodiment of the present invention,
10 is a seat, and 11 is a first microphone.

12 is a second microphone group installed on the backrest of the seat 10, 13 is a pressure sensor group installed on the backrest 16 of the seat 10, 14 is a signal selection section, and 16 is a voice recognition means. The first microphone 11 is used to collect the voice of the speaker, and the second microphone group 12 is used to detect the presence or absence of the voice of the speaker sitting on the seat 1o. Moreover, the pressure sensor group 13 corresponds to the speaker sitting in the seat 1o.

It is used to detect the contact portion with the backrest portion 16 of the seat 10.

An outline of the operation when the speaker speaks while sitting on the seat 10 will be explained below. The first microphone 11 inputs the voice of the speaker to the voice recognition means 16 . The second microphone group 12 collects signals accompanying the utterances of the speaker and inputs them to the signal selection section 14 . On the other hand, the pressure sensor group 13 inputs a signal corresponding to the contact state of the speaker to the signal selection section 14.

The signal selection unit 14 performs processing to detect the contact portion between the speaker and the seat 10 based on the signal from the pressure sensor group 13, and selects the microphone in the second microphone group 12 that is in contact with the speaker. , inputs the signal of the selected microphone to the voice recognition means 16. The speech recognition means 16 performs speech recognition while controlling the determination of the segment of the speech of the human input signal picked up by the first microphone 11 using the signal from the signal selection section 14 .

Here, the signal from the signal selection unit 14 is a signal from the second microphone group 12 that is in contact with the speaker. Since the information is collected while being separated from the external noise environment between the person and the seat 10, the presence or absence of the voice of the speaker is not affected by ambient noise. Therefore, the cutout section of the audio signal of the first microphone 11 can be determined accurately regardless of the surrounding sounds, and the recognition rate of speech recognition can be improved.

Note that the second microphone group 12 and the pressure sensor group 13
The positional relationship with and the number are arbitrary. When the second microphone group 12 and the pressure sensor group 13 are arranged as a pair, the processing 17Ij of the signal selection section 14 becomes easy, and when they are located at the same position, the second microphone group 12 The accuracy of contact detection between the seat 10 and the seat 10 increases. Furthermore, it is easily possible to integrate the pressure sensor group 13 and the second microphone group 12.

Effects of the Invention According to the present invention, it is possible to provide a speech recognition device that has good speech recognition characteristics when a speaker is sitting on a seat in the presence of four sounds in the surroundings. Therefore, great effects can be obtained when applied to speech recognition devices in moving bodies with strong ambient noise, such as cars, trains, and airplanes.

In addition, using a pressure sensor group, it is possible to always detect which microphone is in contact with the speaker from the second group of microphones installed in the seat, so the breathing condition of the speaker can always be monitored under good conditions, and the voice can be monitored. Not only can the cutout section be determined more accurately, but it can also be applied as a pickup to detect whether the speaker is asleep.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a diagram schematically showing a speech recognition device according to an embodiment of the present invention, and FIG. 2 is a diagram showing a conventional example (1/). 10... Seat, 11... First microphone, 12... Second microphone group, 13
. . . Pressure sensor group, 14 . . . Signal selection section, 15 . . . Voice recognition means 5,

Claims (1)

[Claims] A first microphone that picks up the speaker's voice, a second group of microphones installed on the seat, a group of pressure sensors installed on the seat, and a second microphone based on the signals from the pressure sensor group.
A signal selection unit that selects an input signal from a group of microphones, and a voice recognition unit that controls a segment of the input audio signal by the first microphone based on the signal from the signal selection unit. Speech recognition device.