US4078154A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Voice recognition system using locus of centroid of vocal frequency spectra
Assignee: Fujifilm Business Innovation Corp
•
Filing: 1976-08-09
•
Granted: 1978-03-07
•
Expires: 1996-08-09
Abstract
Abstract A voice recognition system is disclosed using the centroid of frequency spectra obtained by frequency analysis of a voice as the characteristic parameter of the voice. The locus of the centroid is obtained by periodically measuring, during the utterance of a selected phrase, the average frequency and the average amplitude. The locus constitutes a pattern of the centroid of frequency spectra and is stored in a memory. During recognition a person keys in a known address to extract the stored pattern of his voice from the memory and then reutters the key phrase. The centroid of freqency spectra is recalculated and the locus obtained is compared with the one in the memory. Humming distance is defined as the distance between a recorded locus of centroid of frequency and a newly inputted locus of centroid of frequency.
Inventors
- 👤 Matsumi Suzuki
- 👤 Yoichi Ichikawa
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-10-24 10:01
CPC Classifications
G; G10; G10L; G10L15/00; G10L15/02
Claims Analysis
Total Claims: 5
Claim 1
Independent Claim
111 words
A voice recognition system comprising first means for periodically detecting the average frequency and average amplitude of at least a first voice signal to obtain a locus of the centroid of frequency spectra of said first voice, means for storing said locus, means for calculating a humming distance between a selected stored locus of the centroid of frequency spectra and a locus of the centroid of frequency spectra of a second voice detected by said first means, means for comparing said humming distance to a preselected value, a humming distance below said preselected value being accepted as an indication that both loci were detected from voice signals from a single person.
Claim 2
Dependent Claim
111 words
A voice recognition system comprising first means for periodically detecting the average frequency and average amplitude of at least a first voice signal to obtain a locus of the centroid of frequency spectra of said first voice, means for storing said locus, means for calculating a humming distance between a selected stored locus of the centroid of frequency spectra and a locus of the centroid of frequency spectra of a second voice detected by said first means, means for comparing said humming distance to a preselected value, a humming distance below said preselected value being accepted as an indication that both loci were detected from voice signals from a single person.
Claim 3
Dependent Claim
95 words
A voice recognition system as claimed in claim 1 wherein said means for periodically detecting comprises, a bank of band pass filters for dividing a voice signal into frequency components f l . . . f n , a bank of peak holding circuits, each connected to detect the peak amplitude during a sample period of the respective components f l . . . f n , and centroid generating means for calculating during each sample period, the average amplitude and the average frequencies based on the peak values obtained by the peak holding circuits.
Claim 4
Dependent Claim
75 words
A voice recognition system as claimed in claim 2 wherein said centroid generating means for calculating comprises first means for calculating the average frequency F t = ΣAifi/ΣAi and second means for calculating the average amplitude A t = ΣAi/n, where n is the number of frequency components f l . . . f n , Ai is the peak value at center frequency fi, and the sum Σ is taken from i-l to i-n.
Claim 5
Dependent Claim
103 words
A voice recognition system as claimed in claim 3 wherein the means for calculating the humming distance is a computer which calculates: ##EQU2## where: l H is the humming distance between a recorded locus of centroid of frequency and a newly imputted locus of centroid of frequency; Fi is the average frequency of the input pattern at the "ith" position; Fi* is the average frequency of the standard or recorded pattern at the "ith" position; Ai is the average amplitude of the input pattern at the "ith" position; Ai* is the average amplitude of the standard or recorded pattern at the "ith" position.
References Cited (Backward Citations)
US3810156A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 1981-03-06 | 1981-09-05 | 1982-03-07 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 1985-03-06 | 1985-09-05 | 1986-03-07 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 1989-03-06 | 1989-09-05 | 1990-03-07 | $7700.00 | Window Closed / Lapsed |