Voice recognition system
First Claim
1. A system for registering a voice pattern by superposing two or more voice patterns for the same sound a multiple of times, comprising:
- converting means for converting a sound into an electrical voice signal;
processing means for processing said electrical voice signal in a predetermined manner to produce a voice pattern in the form a time-frequency distribution;
detecting means for detecting a section from the beginning of said voice pattern where voice energy of said sound is equal to or less than a first predetermined value and partial voice energy of a low frequency component of said sound is equal to or larger than a second predetermined value; and
superposing means for superposing said voice pattern onto a previously created voice pattern for the same sound after completion of said section to form a combined voice pattern.
1 Assignment
0 Petitions
Accused Products
Abstract
A voice or sound recognition system including a microphone for converting a voice into an electrical voice signal, a frequency analyzer for generating a voice pattern in the form of a time-frequency distribution, and a matching unit for matching the voice pattern with registered voice patterns. A voice pattern sometimes contains a bass bar section starting from the beginning of the voice pattern over a time period. In one form of the present invention, the bass bar section is detected and/or eliminated before processing the voice pattern for the purpose of registration or matching. The voice level is sometimes too strong or too weak for appropriate processing. In another form of the present invention, a voice recognition system including a voice input cancelling function is provided. In addition, a system for detecting a voice interval for use in voice data processing, including a device for determining a start point of the voice interval is provided.
31 Citations
40 Claims
-
1. A system for registering a voice pattern by superposing two or more voice patterns for the same sound a multiple of times, comprising:
-
converting means for converting a sound into an electrical voice signal; processing means for processing said electrical voice signal in a predetermined manner to produce a voice pattern in the form a time-frequency distribution; detecting means for detecting a section from the beginning of said voice pattern where voice energy of said sound is equal to or less than a first predetermined value and partial voice energy of a low frequency component of said sound is equal to or larger than a second predetermined value; and superposing means for superposing said voice pattern onto a previously created voice pattern for the same sound after completion of said section to form a combined voice pattern. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8)
-
-
9. A system for registering a voice pattern by superposing two or more voice patterns for the same sound a multiple of times, comprising:
-
converting means for converting a sound into an electrical voice signal; processing means for processing said electrical voice signal in a predetermined manner to produce a voice pattern in the form of a time-frequency distribution. detecting means for detecting a section from the beginning of said voice pattern where voice energy of said sound is equal to or less than a first predetermined value and partial voice energy of a low frequency component of said sound is equal to or larger than a second predetermined value, said detecting means cutting off said section upon detection; and superposing means for superposing said voice pattern onto a previously created voice pattern for the same sound to form a combined voice pattern. - View Dependent Claims (10, 11, 12, 13, 14, 15, 16)
-
-
17. A voice recognizing system, comprising:
-
converting means for converting a voice to be recognized into an electrical voice signal; processing means for processing said electrical voice signal in a predetermined manner to produce a voice pattern in the form of a time-frequency distribution; detecting means for detecting a section from the beginning of said voice pattern where voice energy of said voice is equal to or less than a first predetermined value and partial voice energy of a low frequency component of said voice is equal to or larger than a second predetermined value; and matching means for matching said voice pattern with at least one of a plurality of registered voice patterns. - View Dependent Claims (18, 19, 20, 21, 22, 23, 24)
-
-
25. A voice recognizing system, comprising:
-
converting means for converting a voice to be recognized into an electrical voice signal; processing means for processing said electrical voice signal in a predetermined manner to produce a voice pattern in the form of a time-frequency distribution and also a voice power signal, said voice pattern being comprised of a collection of frames in timed sequence; identifying means for identifying said voice pattern by matching with a plurality of registered voice patterns; detecting means for detecting a voice interval by comparing said voice power signal with a first reference value; adding means for adding voice powers of said voice power signal over a predetermined number of frames; and means for comparing said added voice power with a pair of lower and upper reference values and cancelling said voice pattern if said added value is outside a range between said lower and upper reference values. - View Dependent Claims (26, 27, 28)
-
-
29. A system for detecting a voice interval for use in voice registration or recognition, comprising:
-
converting means for converting a voice into an electrical voice signal; processing means for processing said electrical voice signal at a predetermined time interval in a predetermined manner to produce a voice pattern in the form of a time-frequency distribution frame by frame, said processing means generating a voice power of said voice pattern, a first partial voice power of a low frequency range of said voice pattern and a second partial voice power of said voice pattern; and comparing means for comparing said first partial voice power with a predetermined threshold value and also with said second partial voice power and determining a start point of said voice interval if said first partial voice power is found to be larger than both of said predetermined threshold value and said second partial voice power. - View Dependent Claims (30, 31, 32, 33)
-
-
34. A system for detecting a voice interval for use in voice registration or recognition, comprising:
-
converting means for converting a voice into an electrical voice signal; processing means for processing said electrical voice signal at a predetermined time interval in a predetermined manner to produce a voice pattern in the form of a time-frequency distribution frame by frame, said processing means generating a voice power of said voice pattern, a first partial voice power of a low frequency range of said voice pattern and a second partial voice power of said voice pattern; and comparing mean for comparing said second partial voice power with a predetermined threshold value and also with said first partial voice power and determining a start point of said voice interval if said second partial voice power is found to be larger than both of said predetermined threshold value and said second partial voice power. - View Dependent Claims (35, 36, 37, 38, 39, 40)
-
Specification