Data processing: speech signal processing – linguistics – language – Speech signal processing – For storage or transmission
Reexamination Certificate
2000-03-20
2003-06-03
Banks-Harold, Marsha D. (Department: 2654)
Data processing: speech signal processing, linguistics, language
Speech signal processing
For storage or transmission
C704S275000, C708S309000
Reexamination Certificate
active
06574592
ABSTRACT:
BACKGROUND OF THE INVENTION
The present invention relates to a voice detecting device for voice detection for the purpose of remote control by predetermined voices to an apparatus normally requiring manual control, and a voice control system using the voice detecting device.
There are voice control systems using voice detection as existing technologies for remote control by voices of apparatuses or appliances normally requiring manual control.
FIG. 1
is a block diagram showing the structure of an existing voice control system.
The existing voice control system includes a microphone
30
for inputting voices and a personal computer
31
for various kinds of processing such as detecting and identifying an input voice, control responsive to the result of identification, and so on.
The personal computer
31
includes a central processing unit (CPU)
32
for various operations, a memory
33
for storing various data, a controlled objective block
34
to be controlled by voice detection, and other block
35
.
For conducting control by voice detection, a plurality of comparison voice data to be compared with input voice data are previously stored in the memory
33
to enable detection and identification of the voices. A part of CPU
32
is used as a voice detection block
32
a
for detecting and identifying an input voice and executing a control responsive to the identification result.
When a voice of a plurality of words or syllables by a speaker is input through the microphone
30
, the voice detection block
32
a
detects the input voice data, and compare it with individual comparison voice data read out from the memory
33
to determine whether there is any that corresponds to it. If there is a comparison voice data which corresponds to the input voice data, a control command (manipulation command) corresponding to the comparison voice data is output to the controlled objective block
34
. The controlled objective block
34
supplied with the control command works according to the control command.
Voice control systems of the above-introduced type need the ability of operation processing of several decade MIPS through several hundred MIPS and memory resources of several decade kB. In large-scale systems such as personal computers, the resources used for voice detection is only a small part of the maximum resources of the entire system, and the said need does not invite enlargement of the systems or increase their costs.
Small-scale systems, however, which are intended only for detection of a few manipulation commands, do not need large-scale voice detection systems as those used in personal computers. In an extreme case, inexpensive voice detection means sufficient to attain only one manipulation are sometimes demanded.
Heretofore, however, in case of detecting voices of words or syllables from unspecified persons, resources required for realization thereof, such as operation processing abilities of CPU, memory capacities, and so forth, were enormous, and it invited a significant increase of the cost and scale of systems for remote control by voices.
SUMMARY OF THE INVENTION
It is therefore an object of the present invention to provide a voice detecting device and a voice control system which are simple in construction and inexpensive by making limitations in property of input voices in terms of frequency, magnitude, length, and so on.
According to the invention, there is provided a voice detecting device comprising:
a voice frequency detector for detecting the frequency of a voice input thereto, then discriminating whether the frequency falls within a preset frequency range for voices to be detected, and outputting the result of the discrimination;
an input signal level detector for detecting the energy level of the input voice, then comparing it to confirm whether the detected energy level exceeds a preset energy level threshold value for voices to be detected, and outputting the result of the comparison;
a voice input judge part responsive to the result of discrimination by the voice frequency detector and the result of comparison by the input signal level detector to judge whether a voice satisfying conditions for voice detection has been input or not and output a first status signal in accordance with the result of the judgement; and
a voice duration measure part for measuring the duration of the first status signal, then comparing it to confirm whether the detected duration falls within a preset range of duration threshold value, and outputting a second status signal in accordance with the result of the comparison, and by providing limitations to detected voices in frequency, magnitude, length and other detecting conditions, it is possible to significantly decrease the enormous amount of operations caused by the dictionary memory and complicated detection algorithms as compared with existing techniques and thereby realize voice control by using simple and light system resources at a low cost. Therefore, it is possible to provide a voice detecting device for remote control by voices from unspecified persons to an apparatus normally requiring manual control with a simple structure at a low cost.
If the device is designed to permit changes of the frequency range of detected voices, energy level threshold value and duration threshold value of detected voices, the detected objective voices can be limited in a range for each purpose. That is, by designing the device to be changeable in frequency range of detected voices relative to frequencies of voices to be detected, detection adaptive to a specific speaker, detection adaptive to genders to a certain level, and so on, are possible. By designing the device to be changeable in level threshold value relative to the magnitude of the voice to be detected, erroneous detection of noise other than voices and detection of low voices below a predetermined level, for example, can be prevented. Furthermore, by designing the device to be changeable in threshold value of the duration of voices to be detected, it is possible to alleviate the processing load of a timer or a counter in the control means, or detect only voices of lengths within a predetermined time and thereby prevent erroneous detection.
The voice frequency detector is preferably a digital filter including a band pass filter block for setting the frequency range of voices to be detected and supplied with a voice, and an energy estimator for identifying a voice from a signal coming through the band pass filter block.
The voice control system according to the invention includes a microphone for introducing a voice, a voice detecting device according to the invention, a control means responsive to one or both of the first status signal and the second status signal to determine whether to output a control command, and a controlled objective which executes an operation according to the control command, and with this structure, the invention can provide an inexpensive voice control system with a simple structure for remote control by voices of unspecified speakers about an apparatus normally controlled manually.
REFERENCES:
patent: 4100370 (1978-07-01), Suzuki et al.
patent: 4296277 (1981-10-01), Daneffel
patent: 5548647 (1996-08-01), Naik et al.
patent: 5910904 (1999-06-01), Uramoto
patent: 6023671 (2000-02-01), Iijima et al.
patent: 6182035 (2001-01-01), Mekuria
patent: 6385548 (2002-05-01), Ananthaiyer et al.
patent: 55-13562 (1980-01-01), None
patent: 61-271781 (1986-12-01), None
patent: 62-171010 (1987-07-01), None
patent: 281327 (1996-10-01), None
Lee et al., “A voice activity detection algorithm for communication systems with dynamically varying background acoustic noise,” May 1998, 48th IEEE Vehicle Technology Conference, vol. 2, pp. 1214 to 1218.
Nankawa Keigo
Saito Masaki
Banks-Harold Marsha D.
Kabushiki Kaisha Toshiba
Lerner Martin
LandOfFree
Voice detecting and voice control system does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Voice detecting and voice control system, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Voice detecting and voice control system will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3139335