close

Вход

Забыли?

вход по аккаунту

JP2000132200

код для вставкиСкачать
Patent Translate
Powered by EPO and Google
Notice
This translation is machine-generated. It cannot be guaranteed that it is intelligible, accurate,
complete, reliable or fit for specific purposes. Critical decisions, such as commercially relevant or
financial decisions, should not be based on machine-translation output.
DESCRIPTION JP2000132200
[0001]
BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an
audio / video device with speech recognition and a speech recognition method.
[0002]
2. Description of the Related Art A speech recognition device is a device for recognizing speech
uttered by a human and outputting necessary commands to the device, and is used as one of
remote controllers. For example, if you say "It's hot" to the air conditioner in the house, it
recognizes it and outputs an instruction to lower the temperature, and if you say "Samui", it
outputs an instruction to raise the temperature to The temperature control of the drive can be
performed without manual operation. Such voice recognition devices are also used in home
appliances, personal computers, and in-vehicle devices mounted on vehicles. In the navigation
device which is one of the in-vehicle devices, when uttering the destination, the recommended
route to the destination is automatically selected and displayed. When the facility name is
uttered, the list of the facilities is displayed. Can. In the case of a vehicle-mounted device, remote
control using a voice recognition device is particularly useful so as not to interfere with the
driver's driving operation.
[0003]
However, in the conventional voice recognition apparatus, when voice is input through the
microphone, the voice recognition apparatus is an audio / video apparatus such as a radio, a
08-05-2019
1
television, a cassette tape, a CD, an MD, etc. When attached to another device attached or placed
near an audio / video device, the audio such as human voice or music output from the audio /
video device will be in the way. Unless the volume was lowered or turned off manually, voice for
remote control could not be input to the voice recognition device.
[0004]
The present invention solves such conventional problems and can input human voice to the voice
recognition device even when sound is output from the speaker of the audio / video device
nearby. It is an object of the present invention to provide an audio / video apparatus with speech
recognition function and a speech recognition method.
[0005]
SUMMARY OF THE INVENTION According to the present invention, among the sounds output
from the speaker of an audio / video device, the present invention prevents speech recognition
by human voice input, for example, several hundred Hz. Means for attenuating the sound in the
frequency band from 1 kHz to several kHz and detecting the sound pressure level of the sound
output from the speaker by the speech recognition device, and the sound pressure level is an
obstacle to speech recognition by human speech input In this case, the control means is provided
with control means for controlling the attenuation means to attenuate the sound in the frequency
band.
This enables human voice input to the voice recognition device while listening to music and the
like on the audio / video device.
[0006]
BEST MODE FOR CARRYING OUT THE INVENTION According to the first aspect of the present
invention, there is provided an audio / video device for outputting sound, and an audio based on
human voice input among sounds output from a speaker of the audio / video device. Attenuation
means for attenuating sound in a specific frequency band that interferes with recognition,
recognizing an input human voice and outputting a predetermined command, and detecting a
sound pressure level of the sound output from the speaker Audio with voice recognition function,
comprising: voice recognition device; and control means for controlling the attenuation means
based on the detected sound pressure level to output sound with the specific frequency band
attenuated from the speaker And / or a video device, and has the effect of enabling human voice
input to the voice recognition device while listening to music or the like with the audio / video
08-05-2019
2
device.
[0007]
The invention according to claim 2 of the present invention comprises a plurality of speakers and
a plurality of attenuation means, and the control unit individually controls the attenuation
amount of each attenuation means. Audio / video devices, for example when speakers are
arranged in front and back, the front speakers mainly interfere with human voice input, so by
attenuating the volume to the front speakers While listening to music or the like from the
speaker of (1), it is possible to input human voice to the voice recognition device.
[0008]
The invention according to claim 3 of the present invention is that the audio / video device is an
on-vehicle audio / video device mounted on a vehicle. It is an apparatus and has the effect ¦
action that this invention is applicable with respect to a vehicle-mounted apparatus.
[0009]
The invention according to claim 4 of the present invention is an audio / video apparatus with a
speech recognition function according to claim 3, characterized in that the speech recognition
apparatus is shared between an audio / video apparatus and another apparatus. For example, in
the case where the in-vehicle audio / video device is configured integrally with the in-vehicle
navigation device, the voice recognition device can share the audio / video device and the
navigation device.
[0010]
The invention according to claim 5 of the present invention is the audio / video apparatus with
voice recognition function according to any one of claims 1 to 4, wherein the specific frequency
band is several hundred Hz to several kHz, and human voice input This function has the effect of
reliably attenuating the frequency band that interferes with the
[0011]
The invention according to claim 6 of the present invention is that when a voice is input to a
voice recognition device while sound is being output from a speaker of an audio / video device
nearby, a specific frequency that interferes with the voice input. This is a speech recognition
method characterized by attenuating the sound in the band, and has the effect of enabling human
speech input to the speech recognition device while listening to music or the like with the audio /
video device.
08-05-2019
3
[0012]
The invention according to claim 7 of the present invention is the speech recognition method
according to claim 6, wherein when the audio / video device has a plurality of channels, the
sound of a specific frequency band is attenuated for each channel. Yes, it has the effect of
attenuating only the sound of the channel that interferes with the voice input.
[0013]
The invention according to claim 8 of the present invention is the speech recognition method
according to claim 6 or 7, wherein the specific frequency band is from several hundred Hz to
several kHz, in which the frequency band which interferes with human speech input is assured. It
has the effect of being able to attenuate.
[0014]
Hereinafter, embodiments of the present invention will be described with reference to the
drawings.
(First Embodiment) FIG. 1 shows the configuration of a car-mounted four-channel audio
apparatus with a speech recognition function according to a first embodiment of the present
invention.
In FIG. 1, 1 is an antenna, 2 is a radio receiver, 3 is a band pass attenuator for attenuating
frequencies in a specific frequency band from several hundred Hz to several kHz, and 4, 5, 6, 7
are amplifiers for each channel , 8, 9, 10, 11 are speakers for each channel, 12 is a microphone,
13 is a voice recognition device, and 14 is a control unit.
The audio device is integrated with the navigation device and disposed in a dashboard in front of
the passenger compartment, and the voice recognition device 13 is shared by the audio device
and the navigation device.
The speakers 8 and 9 are disposed on the left and right sides on the front side of the vehicle
interior, and the speakers 10 and 11 are disposed on the left and right sides on the rear side of
08-05-2019
4
the vehicle interior.
Also, the microphone 12 is embedded in the wheel handle.
[0015]
Next, the operation of the present embodiment will be described with reference to the flow of
FIG.
The FM radio waves of the radio received through the antenna 1 are tuned and detected by the
radio receiver 2, are output as audio signals of four channels, are amplified by the amplifiers 4, 5,
6, 7, respectively, and the speakers 8, 9, 10 and 11 (step S1).
At this time, the microphone 12 converts the sound output from each of the speakers 8 to 11
into an electric signal and inputs the electric signal to the voice recognition device 13.
In this state, when the driver performs voice input through the microphone 12 to change the
operation of the audio device or the navigation device (step S2), the voice recognition device 12
recognizes that the voice signal input from the driver is clear If it can not be recognized (step S3)
and the sound signals input from the speakers 8 to 11 interfere and the sound signal input from
the driver can not be clearly recognized, an error signal is sent to the control unit 14 Output.
The control unit 14 attenuates the frequency of the specific frequency band from several
hundred Hz to several kHz in the band pass attenuator 3 based on the error signal (step S4). As a
result, since the speakers 8 to 11 output the sound in which the frequency of this specific
frequency band is attenuated, the sound signal in this band is only the sound signal input by the
driver through the microphone 12, The recognition device 13 can recognize the speech clearly.
[0016]
As described above, according to the first embodiment, while the sound recognition device 13
detects the sound pressure level of the sound output from the speakers 8 to 11, for example, the
music through the microphone 12, the driver operates the microphone 12 in this state. If the
voice input through the speaker can not be clearly recognized, the control unit 14 controls the
08-05-2019
5
band pass attenuator 3 so that the frequency of the specific band from several hundred Hz to
several kHz from the frequency of the music output from the speakers 8 to 11 Since the
attenuation is performed, the driver can perform voice input to the voice recognition device 13
while listening to music in another frequency band.
[0017]
(Second Embodiment) FIG. 3 shows the configuration of an on-vehicle four-channel audio
apparatus with a voice recognition function according to a second embodiment of the present
invention, and the first embodiment shown in FIG. The band pass attenuator 3 is divided into four
channels, and the control unit 14 can individually control the band pass attenuators 3a, 3b, 3c,
and 3d.
[0018]
Therefore, in the second embodiment, while the sound recognition device 13 detects the sound
pressure level of the sound output from each of the speakers 8 to 11, for example, music through
the microphone 12, the driver inputs it through the microphone 12 in this state. If the voice can
not be recognized clearly, the control unit 14 controls only the bandpass attenuators 3a and 3b,
and several hundred Hz to several kHz from the frequency of the music output from the front
speakers 8 and 9 near the microphone 12. By attenuating the frequency of the specific band up
to the point, the driver can perform voice input to the voice recognition device 13 while listening
to the music from the unattenuated rear speakers 10 and 11.
[0019]
In each of the above embodiments, the voice recognition device 13 detects the recognition rate of
the driver's voice input through the microphone 12 and outputs it to the control unit 14, and the
control unit 14 uses the recognition rate as input. Accordingly, the attenuation factor in the band
pass attenuator may be controlled.
Also, the speech recognition device 13 may be dedicated to an audio / video device.
Furthermore, the audio / video device may be for household use other than in-vehicle use.
[0020]
08-05-2019
6
As described above, according to the present invention, among the sounds output from the
speaker of the audio / video device, for example, from several hundred Hz to several kHz, which
interfere with speech recognition by human speech input. In addition to the means for
attenuating the sound in the frequency band, when the sound pressure level of the sound output
from the speaker is detected by the speech recognition device, and the sound pressure level
interferes with the speech recognition by human speech input, The control means for controlling
the attenuation means to attenuate the sound in the above-mentioned frequency band has the
effect of enabling human voice input to the voice recognition device while listening to music or
the like with the audio / video device.
08-05-2019
7
1/--страниц
Пожаловаться на содержимое документа