EP0071716A2
Expired / Lapsed
EP
⏱️ 0.0 years remaining
Allophone vocoder
Assignee: Texas Instruments Inc
•
Filing: 1982-06-14
•
Granted: —
•
Expires: 2002-06-14
Abstract
Abstract An allophone vocoder which utilizes the inherent redundancy of the spoken language together with the automatic human filtering of speech so as to obtain a speech compression and recognition system. An analog speech signal is broken up into its phoneme components (105) and encoded for transmission (110). The encoded phoneme sequence has a much higher compression rate than the analog speech signal. The phonemes are then either transmitted (110), stored (109), or used to generate directly an analogous allophone sequence (114) so as to approximate the original speech signal. Due to the inherent redundancy of the spoken language, and the filtering effect of the human, variations or errors in the approximations of the phonemes received from the original speech signal are inconsequential to the comprehension ability of the final allophone synthesized speech.
Inventors
- 👤 Granville E. Ott
- 👤 Kun-Shan Lin
Bibliographic Details
Application Number
—
Jurisdiction
EP
Data Source
google_patents
Last Synchronized
2026-11-01 03:39
CPC Classifications
G; G10; G10L; G10L19/00; G10L19/0018
Claims Analysis
Total Claims: 13
Claim 1
Independent Claim
61 words
A recognition system for receiving analog speech signals comprising; a. means, 1. for separating said analog signal into human sound component parts, and 2. for generating perceived indicia representative of each of said human sound component parts; and b. means for selecting for each human sound component part, based on said perceived indicia a reference data from reference a data set.
Claim 2
Dependent Claim
61 words
A recognition system for receiving analog speech signals comprising; a. means, 1. for separating said analog signal into human sound component parts, and 2. for generating perceived indicia representative of each of said human sound component parts; and b. means for selecting for each human sound component part, based on said perceived indicia a reference data from reference a data set.
Claim 3
Dependent Claim
31 words
The recognition system according to claim 1, further comprised of means .1. for receiving said analog speech signal, and 2. for transmitting said analog speech signal to said means for separating.
Claim 4
Dependent Claim
18 words
The recognition system according to claim 1 wherein said reference data has reference indicia approximating said perceived indicia.
Claim 5
Independent Claim
106 words
A device for compression of analog speech data into equivalent digital data comprised of: a. means for translating said analog speech data into electronic signals; b. means for separating said electronic signal into a sequence of component human sound; c. means for assigning each of said component human sound a representative indicia; d. means for determining, for each of said representative indicia, a selected human sound from a reference set of human sound, said selected human sounds having indicia being the closest approximation to said representative indicia within said reference set of human sounds; and e. means for digitally storing each of said selected human sounds.
Claim 6
Independent Claim
106 words
A device for compression of analog speech data into equivalent digital data comprised of: a. means for translating said analog speech data into electronic signals; b. means for separating said electronic signal into a sequence of component human sound; c. means for assigning each of said component human sound a representative indicia; d. means for determining, for each of said representative indicia, a selected human sound from a reference set of human sound, said selected human sounds having indicia being the closest approximation to said representative indicia within said reference set of human sounds; and e. means for digitally storing each of said selected human sounds.
Claim 7
Dependent Claim
110 words
A vocoder system comprised of: a. means 1. for accepting an analog speech signal, and 2. for identifying allophone parts of said analog speech signal, and 3. for determining, for each of the allophone parts, a representative indicia; b. means 1. for determining, for each indicia, an optimal comparision value, each optimal comparision value being representative of the similiarity between said representative indicia and reference indicia within a library of reference indicia, and 2. for determining, for each optimal comparision value, an associated allophone code; and c. means 1. for creating an allophone sequence of said associated allophone codes, and 2. for synthesizing an auditory signal from said allophone sequence.
Claim 8
Dependent Claim
110 words
A vocoder system comprised of: a. means 1. for accepting an analog speech signal, and 2. for identifying allophone parts of said analog speech signal, and 3. for determining, for each of the allophone parts, a representative indicia; b. means 1. for determining, for each indicia, an optimal comparision value, each optimal comparision value being representative of the similiarity between said representative indicia and reference indicia within a library of reference indicia, and 2. for determining, for each optimal comparision value, an associated allophone code; and c. means 1. for creating an allophone sequence of said associated allophone codes, and 2. for synthesizing an auditory signal from said allophone sequence.
Claim 9
Dependent Claim
78 words
A formant tracker receiving an analog signal, said formant tracker comprising; a. means for tracking a pitch associated with said analog signal and yielding a fundamental frequency therefrom; and b. means for iteratively establishing for each selected frequency range within a set of selected frequency ranges, 1. an integer set associated with the selected frequency range and 2. an optimal integer for said frequency range from said integer set, and 3. a bandwidth associated with said optimal integer.
Claim 10
Dependent Claim
78 words
A formant tracker receiving an analog signal, said formant tracker comprising; a. means for tracking a pitch associated with said analog signal and yielding a fundamental frequency therefrom; and b. means for iteratively establishing for each selected frequency range within a set of selected frequency ranges, 1. an integer set associated with the selected frequency range and 2. an optimal integer for said frequency range from said integer set, and 3. a bandwidth associated with said optimal integer.
Claim 11
Dependent Claim
69 words
The formant tracker according to claim 6 wherein said means for establishing an optimal integer includes; a. means for iteratively generating, for each integer within said integer set, 1. a sinusoidal signal around a chosen center point, and 2. a convolution value being the convolution of said sinusoidal and said analog signal over said frequency range; and b. means for choosing the optimal integer based on said convolution value.
Claim 12
Dependent Claim
50 words
A vocoder network receiving an analog speech signal, said vocoder network comprising: a. means for recognizing a sequence of phonemes within said analog speech signal; and b. means, 1. for translating said phoneme sequence into an analogous allophone sequence, and 2. for synthesizing said allophone sequence into an auditory signal.
Claim 13
Dependent Claim
50 words
A vocoder network receiving an analog speech signal, said vocoder network comprising: a. means for recognizing a sequence of phonemes within said analog speech signal; and b. means, 1. for translating said phoneme sequence into an analogous allophone sequence, and 2. for synthesizing said allophone sequence into an auditory signal.
References Cited (Backward Citations)
No citation records indexed.
USPTO Annuity & Maintenance Windows
Maintenance schedule is calculated for granted US utility patents.